The standard edition of the complete psychological works of Sigmund Freud. Vol. 13, Totem and taboo ; and other works : (1913-1914) 9780099426660, 0099426668

Citation preview

THE STANDARD EDITION OF THE COMPLETE PSYCHOLOGICAL WORKS OF SIGMUND FREUD

* VOLUME XIII

MICHELANGELO'S MOSES

THE STANDARD EDITION OF THE COMPLETE PSYCHOLOGICAL WORKS OF

SIGMUND FREUD Translated from the German under the General Editorship of

JAMES S T RAC HE Y In Collaboration with

A NNA FREUD Assisted by ALIX STRACHEY and ALAN TYSON

VOLUME XIII

(1913-1914)

Totem and Taboo and

Other Works

LON DON

THE HOG ARTH PRESS AND THE INSTITUTE OF PSYCHO�ANALYSIS

P UB L IS HED BY THE HOGARTH P R ES S L I MI T ED 'TOTEM A N D T AB O O' I S I NCLU D ED BY ARR A N GEMENT WITH ROUTLEDGE A N D KEGA N P AUL LTD. LONDON

*

CLARKE, IRWI N A N D CO. LTD. TORONTO

This Edition first Published in 1955 Reprinted with Corrections 1958 Reprinted 1962, 1964, 1968, 1971, 1973, 1975, 1978 and 1981

I SBN O 7012 0067 7 itm:!:IQJ,QSil2

.BF 173

pgc,

All rights ,,E�pu.frtt�cation may be reproduced, stored in a retrieval system, or transmitted, in any form, or by any means, electronic, mechanical, photo­ copying, recording or otherwise, without the prior permission of The Hogarth Press Ltd.

TRANSLATION AND EDITORI A L MATTER @ THE INSTITUTE OF P SYCHO-AN A LY S I S AND ANGELA RICHARDS I 95:5

PRINTED AND B O UND IN GREAT BRITA IN BY BUTLER AND TANN ER LTD, FROME

R{l L

II II

II

CONTENTS VOLUME THIRTEEN

TOTEM AND TABOO (1913 [1912-13]) page ix Editor's Note xiii Preface xv Preface to the Hebrew Translation l I. The Horror of Incest 18 II. Taboo and Emotional Ambivalence 75 III. Animism, Magic and the Omnipotence of Thoughts 100 IV. The Return of Totemism in Childhood APPENDIX: List of Writings by Freud dealing with Social Anthropology, Mythology and the History of Religion

I

p

162

THE CLAIMS OF PSYCHO-ANALYSIS TO SCIENTIFIC INTERES! (1913) Part I.

The Psychological Interest of Psycho-Analysis

165

Part II.

The Claims of Psycho-Analysis to the Interest of the Non-Psychological Sciences (A) The Philological Interest of Psycho-Analysis (B) The Philosophical Interest of Psycho-Analysis (c) The Biological Interest of Psycho-Analysis (o) The Interest of Psycho-Analysis from a Develop­ mental Point of View (E) The Interest of Psycho-Analysis from the Point of View of the History of Civilization (F) The lnterest,of Psycho-Analysis from the Point of View of the Science of Aesthetics (o) The Sociological Interest of Psycho-Analysis (H) The Educational Interest of Psycho-Analysis

� � OBSERVATIONS AND EXAMPLES FROM co ANALYTIC PRACTICE (1913)

0 -' ,-:,

V

176 178 179 182

184 187

188

189

193

�

CONTENTS F AUSSE RECONNAISSANCE ('DEJA RACONTE') page IN PSYCHO-ANALYTIC TREATMENT (1914) 201

THE MOSES OF MICHELANGELO (1914) Postscript (1927) SOME REFLECTIONS ON SCHOOLBOY PSYCHOLOGY (1914)

211 237 241 245

BIBLIOGRAPHY AND AUTHOR INDEX LIST OF ABBREVIATIONS GENERAL INDEX

254 255

LIST OF ILLUSTRATIONS Michelangelo's Moses

frontispiece

Detail of Michelangelo's Moses

facing page 223

Statuette of Moses

facing page 237

By permission of the Ashmolean Museum, Oxford

TOTEM AND TABOO Some Points of Agreement between the Mental Lives of Savages. and Neurotics (1913 [1912-13])

EDITOR'S NOTE TOTEM UND TABU

(a) GERMAN EDITIONS: 1912 Part I, Imago, 1 (1), 17-33. (Under the title'O'ber einige O'bereinstimmungen im Seelenleben der Wilden und der Neurotiker' ['Some Points of Agreement between the Mental Lives of Savages and Neurotics'].) 1912 Part II, Imago, 1 (3), 213-27 and (4), 301-33. (Same title.) 1913 Part III, Imago, 2 (1), 1-21. (Same title.) 1913 Part IV, Imago, 2 (4), 357-408. (Same title.) 1913 In one volume, under the title Totem und Tahu, Leipzig and Vienna: Heller. Pp. v + 149. 1920 2nd ed. Leipzig, Vienna and Zurich: Internationaler Psychoanalytischer Verlag. Pp. vii + 216. 1922 3rd ed. Leipzig, Vienna and Zurich: I.P.V. Pp. vii + 216. 1924 G.S., 10, 3-194. 1934 5th ed. Vienna: I.P.V. Pp. 194. 1940 G. W., 9. Pp. 1-205. 1934 'Vorrede zur hebraischen Ausgabe von Totem und Tahu.' G.S., 12, 385. 1948 G. W., 14, 569.

(b)

ENGLISH TRANSLATIONS:

Totem and Taboo

1918 New York: Moffat, Yard. Pp. xi + 265. (Tr. A. A. Brill.) 1919 London: Routledge. Pp. xi + 265. (Tr. A. A. Brill.) 1938 London and New York: Penguin Books. Pp. 159. (Tr. A. A. Brill.)

x

TOTEM AND TABOO

1938 In The Basic Writings of Sigmund Freud. New York: Modern Library. Pp. 807-930. (Tr. A.A. Brill.) 1950 London: Routledge and Kegan Paul.Pp. xi + 172. (Tr. James Strachey.)

1950 'Preface to the Hebrew Translation of Totem and Taboo.' In Totem and Taboo, London, 1950, p. xi. (Tr.James Strachey.)

The present translation is a slightly corrected version of the one published in 1950. In his Preface Freud tells us that his first stimulus for writing these essays came from the works of Wundt and Jung.Actually, of course, his interest in social anthropology went back much further. In the Fliess correspondence (1950a), apart from general allusions to his long-standing devotion to the study of archaeology and prehistory, there are a number of specific references to anthropological topics and to the light which psycho-analysis throws upon them. For instance, in Draft N (May 31, 1897) in discussing the 'horror of incest' he touched upon the relation between the growth of civilization and the suppression of the instincts-a subject to which he returned in his paper on '"Civilized" Sexual Ethics' (1908d) and, much later, in Civilization and its Discontents (1930a). Again, in Letter 78 (Dec. 12, 1897) he writes: 'Can you imagine what "endo­ psychic myths" are? They are the latest offspring of my mental labours. The dim inner perception of one's own psychical apparatus stimulates illusions of thought, which are naturally projected outwards and characteristically into the future and the world beyond. Immortality: retribution, life after death, are all reflections of our inner psyche ... psycho-mythology.' And, in Letter 144 (July 4, 1901): 'Have you read that the English have excavated an old palace in Crete (Knossos) which they declare is the authentic labyrinth of Minos? Zeus seems originally to have been a bull. It seems, too, that our own old God, before the sublimation instigated by the Persians took

EDITOR'S NOTE

xi

place, was also worshipped as a bull. That provides food for all sorts of thoughts which it is not yet time to set down on paper.' Lastly it is worth mentioning a short passage in a footnote to the first edition of The Interpretation of Dreams (1900a), near the end of Section B of Chap. V, Standard Ed., 4, 217n., which adumbrates the derivation of the monarchy from the social position of the father of the family. But the major elements of Freud's contribution to social anthropology made their first appearance in this work, and more especially in the fourth essay, which contains his hypo­ thesis of the primal horde and the killing of the primal father and elaborates his theory tracing from them the origins of almost the whole of later social and cultural institutions. Freud himself had a very high opinion of this last essay both as regards its content and its form. He told his present translator, prob­ ably in 1921, that he regarded it as his best-written work. Nevertheless, Dr. Ernest Jones informs us that as late as the middle of June 1913, when the essay was already in proof and after he had presented it before the Vienna Psycho-Analytical Society, he was still expressing doubts and hesitations about publishing it. These doubts were soon removed, however, and the book remained a favourite all through his life and he con­ stantly recurred to it. For instance, he summarized and dis­ cussed it with particular care in the sixth chapter of his Auto­ biographical Study ( 1925d), and he quoted it many times in his last published volume, Moses and Monotheism (1939a). About the actual composition of these essays we have a good deal of information, details of which will be found in the second volume of Dr. ErnestJones's biography of Freud. He had begun his preparations for the work, and in particular his reading of a large amount of literature on the subject, as early as in 1910. The title 'Totem and Taboo' was evidently already in his mind in August, 1911, though he did not finally adopt it till the essays were collected in volume form. The first essay was finished in mid-January, 1912. It was published in Imago in the following March, and was shortly afterwards reprinted, with some small omissions, in the Vienna weekly journal Pan (April 11 and 18, 1912) and in the Vienna daily paper Neues Wiener Journal (April 18). The second essay was given before the Vienna

xii

TOTEM AND TABOO

Psycho-Analytical Society on May 15, 1912, in a talk which lasted for three hours. The third was prepared during the autumn ofl 912 and given before the Vienna Society onJanuary 15, 1913. The fourth was finished on May 12, 1913, and given before the Vienna Society on June 4, 1913. Totem and Taboo was translated into several languages besides English during Freud's lifetime: into Hungarian (1919), Spanish (1923), Portuguese (n.d.), French (1924), Japanese (twice, 1930 and 1934), and Hebrew (1939). For the last of these, as will be seen below (p. xv), he wrote a special preface.

PREFACE THE four essays that follow were originally published (under a heading which serves as the present book's sub-title) in the first two volumes of Imago, a periodical issued under my direction. They represent a first attempt on my part at applying the point of view and the findings of psycho-analysis to some unsolved problems of social psychology [Volkerpsychologie]. Thus they offer a methodological contrast on the one hand to Wilhelm Wundt's extensive work, which applies the hyp otheses and working methods of non-analytic psychology to the same purposes, and on the other hand to the writings of the Zurich school of psycho­ analysis, which endeavour, on the contrary, to solve the prob­ lems of individual psychology with the help of material derived from social psychology. (C£ Jung, 1912 and 1913.) I readily confess that it was from these two sources that I received the first stimulus for my own essays. I am fully conscious of the deficiencies of these studies. I need not mention those which are necessarily characteristic of pioneering work; but others require a word of explanation. The four essays collected in these pages aim at arousing the interest of a fairly wide circle of educated readers, but they cannot in fact be understood and appreciated except by those few who are no longer strangers to the essential nature of psycho-analy­ sis. They seek to bridge the gap between students of such sub­ jects as social anthropology, philology and folklore on the one hand, and psycho-analysts on the other. Yet they cannot offer to either side what each lacks-to the former an adequate initiation into the new psychological technique or to the latter a sufficient grasp of the material that awaits treatment. They must therefore rest content with attracting the attention of the two parties and with encouraging a belief that occasional co-operation between them could not fail to be of benefit to research. It will be found that the two principal themes from which the title of this little book is derived-totems and taboos-have not received the same treatment. The analysis of taboos is put S.F. XIII-B

xiii

-

xiv

PREFACE

forward as an assured and exhaustive attempt at the solution of the problem. The investigation of totemism does no more than declare that 'here is what psycho-analysis can at the moment contribute towards elucidating the problem of the totem'. The difference is related to the fact that taboos still exist among us. Though expressed in a negative form and directed towards another subject-matter, they do not differ in their psychological nature from Kant's 'categorical imperative', which operates in a compulsive fashion and rejects any conscious motives. Totem­ ism, on the contrary, is something alien to our contemporary feelings-a religio-social institution which has been long aban­ doned as an actuality and replaced by newer forms. It has left only the slightest traces behind it in the religions, manners and customs of the civilized peoples of to-day and has been subject to far-reaching modifications even among the races over which it still holds sway. The social and technical advances in human history have affected taboos far less than the totem. An attempt is made in this volume to deduce the original mean­ ing of totemism from the vestiges remaining of it in childhood­ from the hints of it which emerge in the course of the growth of our own children. The close connection between totems and taboos carries us a step further along the path towards the hypo­ thesis presented in these pages; and if in the end that hypo­ thesis bears a highly improbable appearance, that need be no argument against the possibility of its approximating more or less closely to the reality which it is so hard to reconstruct. ROME, Sfptember rgr3

PREFACE TO THE HEBREW TRANSLATION 1 No reader of [the Hebrew version of] this book will find it easy to put himself in the emotional position of an author who is ignorant of the language of holy writ, who is completely es­ tranged from the religion of his fathers-as well as from every other religion-and who cannot take a share in nationalist ideals, but who has yet never repudiated his people, who feels that he is in his essential nature a Jew and who has no desire to alter that nature. If the question were put to him: 'Since you have abandoned all these common characteristics of your countrymen, what is there left to you that is Jewish?' he would reply: 'A very great deal, and probably its very essence.' He could not now express that essence clearly in words; but some day, no doubt, it will become accessible to the scientific mind. Thus it is an experience of a quite special kind for such an author when a book of his is translated into the Hebrew lan­ guage and put into the hands of readers for whom that historic idiom is a living tongue: a book, moreover, which deals with the origin of religion and morality, though it adopts no Jewish standpoint and makes no exceptions in favour of Jewry. The author hopes, however, that he will be at one with his readers in the conviction that unprejudiced science cannot remain a stranger to the spirit of the new Jewry. VIENNA,

December

1930

1 [This preface was first published in German in G.S., 12,385 (1934). It was then stated that a Hebrew translation. was about to be published in Jerusalem by Stybel. Actually it was not published there until 1939, by Kirjeith Zefer.] xv

TOTEM AND TABOO I

THE HOR ROR OF INCEST PREHISTORIC man, in the various stages of his development, is known to us through the inanimate monuments and implements which he has left behind, through the information about his art, his religion and his attitude towards life which has come to us either directly or by way of tradition handed down in legends, myths and fairy tales, and through the relics of his mode of thought which survive in our own manners and customs. But apart from this, in a certain sense he is still our contemporary. There are men still living who, as we believe, stand very near to primitive man, far nearer than we do, and whom we therefore regard as his direct heirs and representatives. Such is our view of those whom we describe as savages or half-savages; and their mental life must have a peculiar interest for us if we are right in seeing in it a well-preserved picture of an early stage of our own development. If that supposition is correct, a comparison between the psy­ chology of primitive peoples, as it is taught by social anthro­ pology, and the psychology of neurotics, as it has been revealed by psycho-analysis, will be bound to show numerous points of agreement and will throw new light upon familiar facts in both sciences. For external as well as for internal reasons, I shall select as the basis of this comparison the tribes which have been described by anthropologists as the most backward and miserable of savages, the aborigines of Australia, the youngest continent, in whose fauna, too, we can still observe much that is archaic and that has perished elsewhere. The Australian aborigines are regarded as a distinct race, showing neither physical nor linguistic relationship with their nearest neighbours, the Melanesian, Polynesian and Malayan 1

2

TOTEM AND TABOO

peoples. They do not build houses or permanent shelters; they do not cultivate the soil; they keep no domesticated animals ex­ cept the dog; they are not even acquainted with the art of mak­ ing pottery. They live entirely upon the flesh of all kinds of animals which they hunt, and upon roots which they dig. Kings or chiefs are unknown among them; communal affairs are decided by a council of elders. It is highly doubtful whether any religion, in the shape of a worship of higher beings, can be attributed to them. The tribes in the interior of the continent, who have to struggle against the hardest conditions of existence as a result of the scarcity of water, appear to be more primitive in all respects than those living near the coast. We should certainly not expect that the sexual life of these poor naked cannibals would be moral in our sense or that their sexual instincts would be subjected to any great degree of restriction. Yet we find that they set before themselves with the most scrupulous care and the most painful severity the aim of avoiding incestuous sexual relations. Indeed, their whole social organization seems to serve that purpose or to have been brought into relation with its attainment. Among the Australians the place of all the religious and social institutions which they lack is taken by the system of 'totemism'. Australian tribes fall into smaller divisions, or clans, each of which is named after its totem. What is a totem? It is as a rule an animal (whether edible and harmless or dangerous and feared) and more rarely a plant or a natural phenomenon (such as rain or water), which stands in a peculiar relation to the whole clan. In the first place, the totem is the common ancestor of the clan; at the same time it is their guardian spirit and helper, which sends them oracles and, if dangerous to others, recognizes and spares its own children. Conversely, the clansmen are under a sacred obligation (subject to automatic sanctions) not to kill or destroy their totem and to avoid eating its flesh (or deriving benefit from it in other ways). The totemic character is inher­ ent, not in some individual animal or entity, but in all the in dividuals of a given class. From time to time festivals are cele­ brated at which the clansmen represent or imitate the motions and attributes of their totem in ceremonial dances. The totem may be inherited either through the female or 4

I. THE HORROR OF INCEST

3

through the male line. It is possible that originally the former method of descent prevailed everywhere and was only subse­ quently replaced by the latter. An Australian's relation to his totem is the basis of all his social obligations: it overrides on the one hand his tribal membership and on the other hand his blood relationships. 1 The totem is not attached to one particular place. The clans­ men are distributed in different localities and live peacefully side by side with members of other totem clans.• And now we come at last to the characteristic of the totemic system which has attracted the interest of psycho-analysts. In 1

'The Totem bond is stronger than the bond ofblood or family in the modern sense.' (Frazer, 1910, 1, 53.) 2 This highly condensed summary of the totemic syl!tem must neces­ sarily be subject to further comments and qualifications. The word 'totem' was first introduced in 1791 (in the form 'totam') from the North American Indians by an Englishman, J. Long. The subject itself has gradually attracted great scientific interest and has produced a copious literature, from which I may select as works ofcapita] importance]. G. Frazer's four-volume Totemism and Exogamy (1910) and the writings of Andrew Lang, e.g. The Secret of the Totem (1905). The merit of having been the first to recognize the importance of totemism for human pre­ history lies with a Scotsman, John Ferguson McLennan (186g-70). Totemic institutions were, or still are, to be observed in operation, not only among the Australians, but also among the North American In­ dians, among the peoples of Oceania, in the East Indies and in a large part ofAfrica. It may also be inferred from certain vestigial remains, for which it is otherwise hard to account, that totemism existed at one time among the Aryan and Semitic aboriginal races of Europe and Asia. Many investigators are therefore inclined to regard it as a necessary phase of human development which has been passed through univer­ sally. How did prehistoric men come to adopt totems? How, that is, did they come to make the fact oftheir being descended from one animal or another the basis oftheir social obligations and, as we shall see presently, of their sexual restrictions? There are numerous theories on the subject -of which Wundt (1906 [264 ff.]) has given an epitome for German readers-but no agreement. It is my intention to devote a special study before long to the problem oftotemism, in which I shall attempt to solve it by the help of a psycho-analytic line of approach. (See the fourth essay in this work.) Not only, however, is the theory of totemism a matter of dispute; the facts themsdves are scarcely capable of being expressed in general

4

TOTEM AND TABOO

almost every place where we find totems we also find a law against persons of the same totem having sexual relations with one another and consequently against their marrying. This, then, is 'exo­

gamy', an institution related to totemism. Strictly enforced as it is, this prohibition is a remarkable one. There is nothing in the concept or attributes of the totem which I have so far mentioned to lead us to anticipate it; so that it is hard to understand how it has become involved in the totemic system. We cannot, therefore, feel surprised that some investi­ gators actually suppose that exogamy had originally-in the earliest times and in its true meaning-nothing to do with totemism, but became attached to it (without there being any underlying connection) at some time when marriage restric­ tions became necessary. However this may be, the bond be­ tween totemism and exogamy exists and is clearly a very firm one. Some further considerations will make the significance of this prohibition clearer: (a) The violation of the prohibition is not left to what might be called the 'automatic' punishment of the guilty parties, as in the case of other totem prohibitions, such as that against killing the totem animal. It is avenged in the most energetic fashion by the whole clan, as though it were a question of averting some danger that threatened the whole community or some guilt that was pressing upon it. A few sentences from Frazer (1910, 1, 54) will show how severely such misdeeds are treated by savages who are otherwise far from being moral by our standards: 'In Australia the regular penalty for sexual intercourse with a terms as I have tried to do in the text above. There is scarcely a state­ ment which does not call for exceptions or contradictions. But it must not be forgotten that even the most primitive and conservative races are in some sense ancient races and have a long past history behind them dur­ ing which their original conditions of life have been subject to much development and distortion. So it comes about that in those races in which totemism exists to-day, we may find it in various stages of decay and disintegration or in the process of transition to other social and reli­ gious institutions, or again in a stationary condition which may differ greatly from the original one. The difficulty in this last case is to decide whether we should regard the present state of things as a true picture of the significant features of the past or as a secondary distortion of them.

I. THE HORROR OF INCEST

5

person of a forbidden clan is death. It matters not whether the woman be of the same local group or has been captured in war from another tribe; a man of the wrong clan who uses her as his wife ii; hunted down and killed by his clansmen, and so is the woman; though in some cases, if they succeed in eluding cap­ ture for a certain time, the offence may be condoned. In the Ta-ta-thi tribe, New South Wales, in the rare cases which occur, the man is killed but the woman is only beaten or speared, or both, till she is nearly dead; the reason given for not actually killing her being that she was probably coerced. Even in casual amours the clan prohibitions are strictly observed; any viola­ tions of these prohibitions "are regarded with the utmost abhor­ rence and are punished by death".' [Quoted from Cameron (1885, 351).] (b) Since the same severe punishment is inflicted in the case of passing love-affairs which have not resulted in any children, it seems unlikely that the reasons for the prohibition are of a practical nature. (c) Since totems are hereditary and not cµanged by mar­ riage, it is easy to follow the consequences of the prohibition. Where, for instance, descent is through the female line, if a man of the Kangaroo totem marries a woman of the Emu totem, all the children, both boys and girls, belong to the Emu clan. The totem regulation will therefore make it impossible for a son of this marriage to have incestuous intercourse with his mother or sisters, who are Emus like himself. 1 (d) But a little more reflection will show that exogamy linked with the totem effects more (and therefore aims at more) than the prevention of incest with a man's mother aud sisters. It makes sl!xual intercourse impossible for a man with all the 1 On the other hand, at all events so far as this prohibition is con­ cerned, the father, who is a Kangaroo, is free to commit incest with his daughters, who are Emus. If the totem descended through the male line, however, the Kangaroo father would be prohibited from incest with his daughters (since all his children would be Kangaroos), whereas the son would be free to commi� incest with his mother. These implications of totem prohibitions suggest that descent through the female line is older than that through the male, since there are grounds for thinking that totem prohibitions were principally directed against the incestuous desires of the son.

6

TOTEM AND TABOO

women of his own clan (that is to say with a number of women who are not his blood-relatives) by treating them all as though they were his blood-relatives. It is difficult at first sight to see the psychological justification for this very extensive restriction, which goes far beyond anything comparable among civilized peopl, since it stresses a characteristic which remains common for all time both to what is sacred and to what is un­ clean: the dread of contact with it. The persistence, however, of this important common characteristic is at the same time evi­ dence that the ground covered by the two was originally one and that it was only as a result of further influences that it be­ came differentiated and eventually developed into opposites. [Ibid., 309.] According to Wundt, this original characteristic of taboo­ the belief in a 'demonic' power which lies hidden in an object and which, if the object is touched or used unlawfully, takes its vengeance by casting a spell over the wrong-doer-is still wholly and solely 'objectified fear'. That fear has not yet split up into the two forms into which it later develops: veneration and horror. [Ibid., 310.] But how did this split take place? Through the transplanting, so Wundt tells us, of the taboo ordinances from the sphere of demons into the sphere of belief in gods. [Ibid., 311.] The con­ trast between 'sacred' and 'unclean' coincides with a succession of two stages of mythology. The earlier of these stages did not completely disappear when the second one was reached but persisted in what was regarded as an inferior and eventually a contemptible form. [Ibid., 312.] It is, he says, a general law of mythology that a stage which has been passed, for the very reason that it has been overcome and driven under by a superior stage, persists in an inferior form alongside the later one, so that the objects of its veneration turn into objects of horror. [Ibid., 3 13.] The remainder of Wundt's discussion deals with the relation of the concept of taboo to purification and sacrifice.

26

TOTEM AND TABOO

(2)

Anyone approaching the problem of taboo from the angle of psycho-analysis, that is to say, of the investigation of the uncon­ scious portion of the individual mind, will recognize, after a moment's reflection, that these phenomena are far from un­ familiar to him. He has come across people who have created for themselves individual taboo prohibitions of this very kind and who obey them just as strictly as savages obey the com­ munal taboos of their tribe or society. If he were not already accustomed to describing such people as 'obsessional' patients, he would find 'taboo sickness' a most appropriate name for their condition. Having learnt so much, however, about this obses­ sional sickness from psycho-analytic examination-its clinical aetiology and the essence of its psychical mechanism-he can scarcely refrain from applying the knowledge he has thus acquired to the parallel sociological phenomenon. A warning must be uttered at this point. The similarity be­ tween taboo and obsessional sickness may be no more than a matter of externals; it may apply only to theforms in which they are manifested and not extend to their essential character. Nature delights in making use of the same forms in the most various biological connections: as it does, for instance, in the appearance of branch-like structures both in coral and in plants, and indeed in some forms of crystal and in certain chemical precipitates. It would obviously be hasty and un­ profitable to infer the existence of any internal relationship from such points of agreement as these, which merely derive from the operation of the same mechanical causes. We shall bear this warning in mind, but we need not be deterred by it from pro­ ceeding with our comparison. The most obvious and striking point of agreement between the obsessional prohibitions of neurotics and taboos is that these prohibitions are equally lacking in motive and equally puzzling in their origin. Having made their appearance at some un­ specified moment, they are forcibly maintained by an irresistible fear. No external threat of punishment is required, for ther� is an internal certainty, a moral conviction, that any violation will

II. TABOO AND EMOTIONAL AMBIVALENCE

27

lead to intolerable disaster. The most that an obsessional patient can say on this point is that he has an undefined feeling that some particular person in his environment will be injured as a result of the violation. Nothing is known of the nature of the in­ jury; and indeed even this wretchedly small amount of informa­ tion is more often obtained in connection with the expiatory and defensive actions which we shall have to discuss later than with the prohibitions themselves. As in the case of taboo, the principal prohibition, the nucleus of the neurosis, is against touching; and thence it is sometimes known as 'touching phobia' or 'delire du toucher'. The prohibition does not merely apply to immediate physical contact but has an extent as wide as the metaphorical use of the phrase 'to come in contact with'. Anything that directs the patient's thoughts to the forbidden object, anything that brings him into intellectual contact with it, is just as much prohibited as direct physical contact. This same extension also occurs in the case of taboo. The purpose of some of the prohibitions is immediately obvi­ ous. Others, on the contrary, strike us as incomprehensible, senseless and silly, and prohibitions of this latter sort are de­ scribed as 'ceremonial'. This distinctions too, is found in the observances of taboo. [See p. 21.] Obsessional prohibitions are extremely liable to displace­ ment. They extend from one object to another along whatever paths the context may provide, and this new object then be­ comes, to use the apt expression of one of my women patients, 'impossible'-till at last the whole world lies under an embargo of 'impossibility'. 'Obsessional patients behave as though the 'impossible' persons and things were carriers of a dangerous in­ fection liable to be spread by contact on to everything in their neighbourhood. I have already [p. 21] drawn attention to the same characteristic capacity for contagion and transference in my description of taboo. We know, too, that anyone who viol­ ates a taboo by coming into contact with something that is taboo becomes taboo himself and that then no one may come into contact with him. I will now put side by side two instances of the transference (or, as it is better to say, the displacement) of a prohibition. One of these is taken from the life of the Maoris and the other from

28

'I

I !

TOTEM AND TABOO

an observation of my own on a female obsessional patient. 'A Maori chief would not blow a fire with his mouth; for his sacred breath would communicate its sanctity to the fire, which would pass it on to the pot on the fire, which would pass it on to the meat in the pot, which would pass it on to the man who ate the meat, which was in the pot, which stood on the fire, which was breathed on by the chief; so that the eater, infected by the chief's breath conveyed through these intermediaries, would surely die.' 1 My patient's husband purchased a household article of some kind and brought it home with him. She insisted that it should be removed or it would make the room she lived in 'impossible'. For she had heard that the article had been bought in a shop situated in, let us say, 'Smith' Street.• 'Smith', however, was the married name of a woman friend of hers who lived in a distant town and whom she had known in her youth under her maiden name. This friend of hers was at the moment 'impossible' or taboo. Consequently the article that had been purchased here in Vienna was as taboo as the friend herself with whom she must not come into contact. Obsessional prohibitions involve just as extensive renunci­ ations and restriciions in the lives of those who are subject to them as do taboo prohibitions; but some of them can be lifted if certain actions are performed. Thereafter, these actions must be performed: they become compulsive or obsessive acts, and there can be no doubt that they are in the nature of expiation, pen­ ance, defensive measures and purification. The commonest of these obsessive acts is washing in water ('washing mania'). Some taboo prohibitions can be replaced in just the same way; or rather their violation can be made good by a similar 'ceremonial'; and here again lustration with water is the preferred method. Let us now summarize the points in which agreement between taboo usages and obsessional symptoms is most clearly shown: ( 1) the f�ct that the prohibitions lack any assignable motive; (2) the fact that they are maintained by an internal necessity; (3) the -fact that they are easily displaceable and that there is a risk of infection from the prohibited object; and (4) the fact that they Frazer (1911b, 136) [quotipg Taylor (1870, 165)]. • ['Hirschengass,' and 'Hirsch' in the original.]

1

II. TABOO AND .EMOTIONAL AMBIVALENCE

29

give rise to injunctions for the performance of ceremonial acts. Now both the clinical history and the psychical mechanism of obsessional neurosis have become known to us through psycho­ analysis. The clinical history of a typical case of 'touching pho­ bia' is as follows. Right at the beginning, in very early child­ hood, the patient shows a strong desire to touch, the aim of which is of a far more specialized kind that one would have been inclined to expect. This desire is promptly met by an exter­ nal prohibition against carrying out that particular kind of touching. 1 The prohibition is accepted, since it finds support from powerful internal forces,• and proves stronger than the in­ stinct which is seeking to express itself in the touching. In con­ sequence, however, of the child's primitive psychical constitu­ tion, the prohibition does not succeed in abolishing the instinct. Its only result is to repress the instinct (the desire to touch) and banish it into the unconscious. Both the prohibition and the in­ stinct persist: the instinct because it has only been repressed and not abolished, and the prohibition because, if it ceased, the in­ stinct would force its way through into consciousness and into actual operation. A situation is created which remains undealt with-a psychical fixation-and everything else follows from the continuing conflict between the prohibition and the instinct. The principal characteristic of the psychological constellation which becomes fixed in this way is what might be described as the subject's ambivalent3 attitude towards a single object, or rather towards one act in connection with that object. He is constantly wishing to perform this act (the touching), [and looks on it as his supreme enjoyment, but he must not perform it] and detests it as well.' The conflict between these two cur­ rents cannot be promptly settled because-there is no other way of putting it-they are localized in the subject's mind in such a manner that they cannot come up against each other. The 1

Both the desire and the prohibition relate to the child's touching his own genitals. 1 That is, from the child's loving relation to the authors of the prohibition. 1 To borrow the apt term coined by Bleuler [1910]. ' [From the Second Edition ( 1920) onwards, the words in square brackets were, perhaps accidentally, omitted.] S,F. Xlll-D

30

TOTEM AND TABOO

prohibition is noisily conscious, while the persistent desire to touch is unconscious and the subject knows nothing of it. If it were not for this psychological factor, an ambivalence like this could neither last so long nor lead to such consequences. In our clinical history of a case we have insisted that the im­ position of the prohibition in very early childhood is the deter­ mining point; a similar importance attaches in the subsequent developments to the mechanism of repression at the same early age. As a result of the repression which has been enforced and which involves a loss of memory-an amnesia-the motives for the prohibition (which is conscious) remain unknown; and all attempts at disposing of it by intellectual processess must fail, since they cannot find any base of attack. The prohibition owes its strength and its obsessive character precisely to its uncon­ scious opponent, the concealed and undiminished desire-that is to say, to an it1ternal necessity inaccessible to conscious in­ spection. The ease with which the prohibition can be transferred and extended reflects a process which falls in with the uncon­ scious desire and is greatly facilitated by the psychological con­ ditions that prevail in the unconscious. The instinctual desire is constantly shifting in order to escape from the impasse and en­ deavours to find substitutes-substitute objects and substitute acts-in place of the prohibited ones. In consequence of this, the prohibition itself shifts about as well, and extends to any new aims which the forbidden impulse may adopt. Any fresh advance made by the repressed libido is answered by a fresh sharpening of the prohibition. The mutual inhibition of the two conflicting forces produces a need for discharge, for reducing the prevailing tension; and to this may be attributed the reason for the performance of obsessive acts. In the case of a neurosis these are clearly compromise actions: from one point of view they are evidences of remorse, efforts at expiation, and so on, while on the other hand they are at the same time substitutive acts t0 compensate the instinct for what has been prohibited. It is a law of neurotic illness that these obsessive acts fall more and more under the sway of the instinct and approach nearer and nearer to the activity which was originally prohibited. Let us now make the experiment of treating taboo as though

II. TABOO AND EMOTIONAL AMBIVALENCE

31

it were of the same nature as an obsessional prohibition in one of our patients. We must make it clear beforehand, however, that many of the taboo prohibitions that come under our notice are of a secondary, displaced and distorted kind, and that we shall have to be satisfied if we can throw only a little light on the most fundamental and significant taboos. Moreover, the differ­ ences between the situation of a savage and of a neurotic are no doubt of sufficient importance to make any exact agreement impossible and to prevent our carrying the comparison to the point of identity in every detail. In the first place, then, it must be said that there is no sense in asking savages to tell us the real reason for their prohibitions­ the origin of taboo. It follows from our postulates that they can­ not answer, since their real reason must be 'unconscious'. We can, however, reconstruct the history of taboo as follows on the model of obsessional prohibitions. Taboos, we must suppose, are prohibitions of primaeval antiquity which were at some time externally imposed upon a generation of primitive men; they must, that is to say, no doubt have been impressed on them violently by the previous generation. These prohibitions must have concerned activities towards which there was a strong in­ clination. They must then have persisted from generation to generation, perhaps merely as a result of tradition transmitted through parental and social authority. Possibly, however, in later generations they may have become 'organized' as an in­ herited psychical endowment. Who can decide whether such things as 'innate ideas' exist, or whether in the present instance they have operated, either alone or in conjunction with educa­ tion, to bring about the permanent fixing of taboos? But one thing would certainly follow from the persistence of the taboo, namely that the original desire to do the prohibited thing must also still persist among the tribes concerned. They must there­ fore have an ambivalent attitude towards their taboos. In their unconscious there is nothing they would like more than to viol­ ate them, but they are afraid to do so; they are afraid precisely because they would like to, and the fear is stronger than the desire. The desire is unconscious, however, in every individual member of the tribe just as it is in neurotics. The most ancient and important taboo prohibitions are the

32

I I

TOTEM AND TABOO

two basic laws of totemism: not to kill the totem animal and to avoid sexual intercourse with members of the totem clan of the opposite sex. These, then, must be the oldest and most powerful of human desires. We cannot hope to understand this or test our hypo­ thesis on these two examples, so long as we are totally ignorant of the meaning and origin of the totemic system. But the word­ ing of these two taboos and the fact of their concurrence will remind anyone acquainted with the findings of psycho-analytic investigations on individuals of something quite definite, which psycho-analysts regard as the centre-point of childhood wishes and as the nucleus of neuroses. 1 The multiplicity of the manifestations of taboo, which have led to the attempts at classification that I have already men­ tioned, are reduced to a single unity by.our thesis: the basis of taboo is a prohibited action, for performing which a strong in­ clination exists in the unconscious. We have heard [see p. 22], though without understanding it, that anyone who does what is forbidden, that is, who violates a taboo, becomes taboo himself. How is this to be brought into line with the fact that taboo attaches not only to a person who has done what is forbidden but also to persons in particular states, to the states themselves, as well as to impersonal objects? What can the dangerous attribute be which remains the same under all these different conditions? There is only one thing it can be: the quality of exciting men's ambivalence and tempting them to transgress the prohibition. Anyone who has violated a taboo becomes taboo himself be­ cause he possesses the dangerous quality of tempting others to follow his example: why should he be allowed to do what is for­ bidden to others? Thus he is truly contagious in that every ex­ ample encourages imitation, and for that reason he himself must be shunned. But a person who has not violated any taboo may yet be per­ manently or temporarily taboo because he is in a state which possesses the quality of arousing forbidden desires in others and of awakening a conflict of ambivalence in them. The majority o� 1 Cf. my forthcoming study upon totemism, to which I have referred more than once in these pages (the fourth essay in this work).

II. TABOO AND EMOTIONAL AMBIVALENCE

33

exceptional positions and exceptional states [see p. 22] are of this kind and possess this dangerous power. The king or chief arouses envy on account of his privileges: everyone, perhaps, would like to be a king. Dead men, new-born babies and women menstruating or in labour stimulate desires by their special helplessness; a man who has just reached maturity stimulates them by the promise of new enjoyment. For that reason all of these persons and all of these states are taboo, since temptation must be resisted. Now, too, we can understand why the amounts of mana pos­ sessed by different persons can be subtracted from one another and can to some extent cancel one another out [see p. 20]. A king's taboo is too strong for one of his subjects because the social difference between them is too great. But a minister may without any harm serve as an intermediary between them. If we translate this from the language of taboo into that of normal psychology, it means something like this. A subject, who dreads the great temptation presented to him by contact with the king, can perhaps tolerate dealings with an official whom he does not need to envy so much and whose position may even seem attain­ able to him. A minister, again, can mitigate his envy of the king by reflecting on the power which he himself wields. So it comes about that smaller differences between the amounts of the tempting magical force possessed by two people are less to be feared than greater ones. It is equally clear why it is that the violation of certain taboo prohibitions constitutes a social danger which must be pun­ ished or atoned for by all the members of the community if they are not all to suffer injury [see p. 20]. If we replace the uncon­ scious desires by conscious impulses we shall see that the danger is a real one. It lies in the risk of imitation, which would quickly lead to the dissolution of the community. If the violation were not avenged by the other members they would become aware that they wanted to act in the same way as the transgressor. We cannot be surprised at the fact that, in the restrictions of taboo, touching plays a part similar to the one which it plays in 'touching phobias', though the secret meaning of the prohibi­ tion cannot be of such a specialized nature in taboo as it is in the neurosis. Touching is the first step towards obtaining any

34

TOTEM AND TABOO

sort of control over, or attempting to make use of, a person or object. We have translated the contagious power inherent in taboo into the possession of some attribute likely to produce tempta­ tion or encourage imitation. This does not appear to tally with the fact that the contagious character of taboo is shown chiefly by its transmissibility on to material objects, which then them­ selves become carriers of taboo. This transmissibility of taboo is a reflection of the tendency, on which we have already remarked, for the unconscious in­ stinct in the neurosis to shift constantly along associative paths on to new objects. Our attention is thus directed to the fact that the dangerous magical force of mana corresponds to two powers of a more realistic sort: the power of reminding a man of his own prohibited wishes and the apparently more important one of inducing him to transgress the prohibition in obedience to those wishes. These two functions can be reduced to one, how­ ever, if we suppose that in a primitive mind the awakening of the memory of a forbidden action is naturally linked with the awakening of an impulse to put that action into effect. Thus recollection and temptation come together again. It must be admitted, too, that, in so far as the example of a man trans­ gressing a prohibition tempts another man to do the same, dis­ obedience to prohibitions spreads like a contagion, in just the same way as a taboo is transferred from a person to a material object and from one material object to another. If the violation of a taboo can be made good by atonement or expiation, which involve the renunciation of some possession or some freedom, this proves that obedience to the taboo injunc­ tion meant in itself the renunciation of something desirable. Emancipation from one renunciation is made up for by the im­ position of another one elsewhere. This leads us to conclude that atonement is a more fundamental factor than purification in the ceremonials of taboo. I will now sum up the respects in which light has been thrown on the nature of taboo by comparing it with the obses­ sional prohibitions of neurotics. Taboo is a primaeval prohibi­ tion forcibly imposed (by some authority) from outside, and

II. TABOO AND EMOTIONAL AMBIVALENCE

35

directed against the most powerful longings to which human beings are subject. The desire to violate it persists in their un­ conscious; those who obey the taboo have an ambivalent atti­ tude to what the taboo prohibits. The magical power that is attributed to taboo is based on the capacity for arousing temp­ tation; and it acts like a contagion because examples are con­ tagious and because the prohibited desire in the unconscious shifts from one thing to another. The fact that the violation of a taboo can be atoned for by a renunciation shows that renun­ ciation lies at the basis of obedience to taboo.

(3) What we now want to discover is how much value is to be attributed to the parallel we have drawn between taboo and obsessional neurosis and to the view of taboo which we have based on that parallel. Their value must clearly depend on whether the view we have put forward has any advantages over others, and whether it gives us a clearer understanding of taboo than we could otherwise reach. We may be inclined to feel that we have given sufficient evidence of the applicability of our view in what has already been said; yet we must attempt to strengthen the evidence by entering into our explanation of taboo prohibitions and usages in greater detail. There is also another path open to us. We can start an in­ quiry as to whether some of the hypotheses which we have car­ ried over from neuroses to taboo or some of the results to which that procedure has led us may not be directly verifiable in the phenomena of taboo. But we must decide what we are to look for. Our assertion that taboo originated in a primaeval prohibi­ tion imposed at one time or other by some external authority is obviously incapable of demonstration. What we shall rather en­ deavour to confirm, therefore, are the psychological determin­ ants of taboo, which we have learnt to know from obsessional neurosis. How did we arrive at our knowledge of these psycho­ logical factors in the case of the neurosis? Through the analyti­ cal study of its symptoms, and particularly of obsessional acts, defensive measures and obsessional commands. We found that they showed every sign of being derived from ambivalent

36

TOTEM AND TABOO

impulses, either corresponding simultaneously to both a wish and' a counter-wish or operating predominantly on behalf of one of the two opposing trends. If, now, we could succeed in demon­ strating that ambivalence, that is, the dominance of opposing trends, is also to be found in the observances of taboo, or '.if we could point to some of them which, like obsessional acts, give simultaneous expression to both currents, we should have established the psychological agreement between taboo and obsessional neurosis in what is perhaps their most important feature. The two fundamental prohibitions of taboo are, as I have al­ ready remarked, inaccessible to our analysis owing to their con­ nection with totemism; while certain others of its injunctions are of a secondary nature and consequently useless for our pur­ pose. For taboo has become the ordinary method of legislation in the communities affected by it and it has come to serve social purposes which are certainly more recent than taboo itself: such, for instance, are the taboos imposed by chiefs and priests for the protection of their own property and privileges. There nevertheless remain a large group of observances on which our investigation can be made. From these I shall select the taboos attaching (a) to enemies, (b) to chiefs and (c) to the dead; and I shall take the material for our examination from the excellent collection included by Frazer in Taboo and the Perils of the Soul (1911b), the second part of his great work The Golden Bough.

(a) The Treatment of Enemies We may be inclined to suppose that savage and half-savage races are guilty of uninhibited and ruthless cruelty towards their enemies. We shall be greatly interested to learn, then, that even in their case the killing of a man is governed by a number of observances which are included among the usages of taboo. These observances fall easily into four groups. They demand (1) the appeasement of the slain enemy, (2) restrictions upon the slayer, (3) acts of expiation and purification by him and (4) certain ceremonial observances. Our incomplete information on the subject does not enable us to determine with certainty how general or the reverse these usages may be among the peoples

II. TABOO AND EMOTIONAL AMBIVALENCE

37

concerned; but for our purposes this is a matter of indifference. It may safely be assumed, in any case, that what we have before us are not isolated peculiarities but widespread usages. The rites of appeasement performed in the island of Timor, when a warlike expedition has returned in triumph bringing the heads of the vanquished foe, are particularly remarkable, since in addition to them the leader of the expedition is sub­ mitted to severe restrictions (see below, p. 39). On the occasion of the expedition's return, sacrifices are offered to appease the souls of the men whose heads have been taken. 'The people think that some misfortune would befall the victor were such offerings omitted. Moreover, a part of the ceremony consists of a dance accompanied by a song, in which the death of the slain man is lamented and his forgiveness is entreated. "Be not angry", they say, "because your head is here with us; had we been less lucky, our heads might now have been exposed in your village. We have offered the sacrifice to appease you. Your spirit may now rest and leave us in peace. Why were you our enemy? Would it not have been better that we should remain friends? Then your blood would not have been spilt and your head would not have been cut off." ' 1 The same is true of the people of Paloo, in Celebes. So, too, 'the Gallas [of East Africa] return­ ing from war sacrifice to the jinn or guardian spirits of their slain foes before they will re-enter their own houses'.1 Other peoples have found a means for changing their former enemies after their death into guardians, friends and benefac­ tors. This method lies in treating their severed heads with affection, as some of the savage races of Borneo boast of doing. When the Sea Dyaks of Sarawak bring home a head from a suc­ cessful head-hunting expedition, for months after its arrival it is treated with the greatest consideration and addressed with all the names of endearment of which their language is capable. The most dainty morsels of food are thrust into its mouth, deli­ cacies of all kinds and even cigars. The head is repeatedly im­ plored to hate its former friends and to love its new hosts since it has now become one of them. It would be a great mistake to 1 1

Frazer (1911b, 166) [quoting Gramberg (1872, 216)]. Frazer (loc. cit.), quoting Paulitschke (1893-6 [2, 50, 136]).

38

TOTEM AND TABOO

suppose that these observances, which strike us as so horrible, are performed with any intention of ridicule. 1 In several of the savage tribes of North America observers have been struck by the mourning over enemies who have been killed and scalped. When a Choctaw had killed an enemy, he went into mourning for a month during which he was subjected to severe restrictions; and the Dacotas had similar practices. When the Osages, reports a witness, have mourned over their own dead, 'they will mourn for the foe just as if he was a friend'.• Before considering the remaining classes of taboo usages in connection with enemies, we must deal with an obvious objec­ tion. It will be argued against us, with Frazer and others, that the grounds for such rites of appeasement are simple enough and have nothing to do with any such thing as 'ambivalence'. These peoples are dominated by a superstitious fear of the ghosts of the slain-a fear which was not unknown in classical antiquity and which was put upon the stage by the great Eng­ lish dramatist in the hallucinations of Macbeth and Richard III. All the rites of appeasement follow logically from this supersti­ tion, as well as the restrictions and acts of expiation which will be discussed presently. This view is also supported by the fourth group of these observances, which can only be explained as attempts at driving away the ghosts of the victims that are pursuing their murderers.8 In addition to this, the savages openly admit their fear of the ghosts of dead enemies and them­ selves attribute to it the taboo usages which we are discussing. This objection is indeed an obvious one, and if it covered the whole ground we could save ourselves the trouble of any further attempt at an explanation. I shall put off dealing with it until later, and for the moment I will merely state the alternative view which is derived from the hypothesis based upon our earlier discussions of taboo. The conclusion that we must draw 1

Frazer (1914, 1, 295), quoting Low (1848 [206]). Frazer (1911b, 181), quoting Dorsey (1884 [126)). 1 Frazer (1911b, 169-74). These ceremonies consist of hf'ating on shields, shouting and screaming, making noises with musical instru­ 1

ments, etc.

II. TABOO AND EMOTIONAL AMBIVALENCE

39 from all these observances is that the impulses which they ex­ press towards an enemy are not solely hostile ones. They are also manifestations of remorse, of admiration for the enemy, and of a bad conscience for having killed him. It is difficult to resist the notion that, long before a table of laws was handed down by any god, these savages were in possession of a living command­ ment: 'Thou shalt not kill', a violation of which would not go unpunished. Let us now return to the other three groups of taboo obser­ vances. Restrictions placed upon a victorious slayer are unusually frequent and as a rule severe. In Timar (cf. the rites of appease­ ment described above, on p. 37) the leader of the expedition is forbidden 'to return at once to his own house. A special hut is prepared for him, in which he has to reside for two months, undergoing bodily and spiritual purification. During this time he may not go to his wife nor feed himself; the food must be put into his mouth by ariother person.' 1 In some Dyak tribes men returning from a successful expedition are obliged to keep to themselves for several days and abstain from various kinds of food; they may not touch iron nor have any intercourse with women. In Logea, an island in the neighbourhood of New Guinea, 'men who have killed or assisted in killing enemies shut themselves up for about a week in their houses. They must avoid all intercourse with their wives and friends, and they may not touch food with their hands. They may eat vegetable food only, which is brought to them cooked in special pots. The in­ tention of these restrictions is to guard the men against the smell of the blood of the slain; for it is believed that if they smelt the blood they would fall ill and die. In the Toaripi or Motumotu tribe of south-eastern New Guinea a man who has killed an­ other may not go near his wife, and may not touch food with his fingers. He is fed by others, and only with certain kinds of food. These observances last till the new moon.' (Frazer, 19nb, 167.) I shall not attempt to give a complete catalogue of the in­ stances quoted by Frazer of restrictions imposed upon victorious manslayers. I will only remark upon a few more such cases in 1 Frazer (1911b, 166), quoting Miiller (1857 [2, 252]).

40

TOTEM AND TABOO

which their taboo character is particularly marked or in which the restrictions are accompanied by expiation, purification and other ceremonials. 'Among the Monumbos of German New Guinea anyone who has slain a foe in war becomes thereby "unclean" '-the same term being applied to women who are menstruating or in child­ bed. He 'must remain a long time in the men's cl '.lb-house, while the villagers gather round him and celebrate his victory with dance and song. He may touch nobody, not even his own wife and children; if he were to touch them it is believed that they would be covered with sores. He becomes clean again by washing and using other modes of purification.' [Ibid., 169.] 'Among the Natchez of North America young braves who had taken their first scalps were obliged to observe certain rules of abstinence for six months. They might not sleep with their wives nor eat flesh; their only food was fish and hasty-pudding . . . . When a Choctaw had killed an enemy and taken his scalp, he went into mourning for a month, during which he might not comb his hair, and if his head itched he might not scratch it ex­ cept with a little stick which he wore fastened to his wrist for the purpose.' [Ibid., 181.] 'When a Pima Indian had killed an Apache, he had to go through severe ceremonies of purification and atonement. Dur­ ing a sixteen-day fast he might not touch meat nor salt, nor look on a blazing fire, nor speak to a human being. He lived alone in the woods, waited on by an old woman, who brought him his scanty dole of food. He bathed often in the river and (as a sign of mourning) kept his head covered with a plaster of mud. On the seventeenth day there was a public ceremony of solemn puri­ fication of the man and his weapons. Since the Pima Indians took the taboo on killing much more seriously than their enem­ ies and did not, like them, postpone the expiation and purifica­ tion till the end of the expedition, their warlike efficiency suf­ fered greatly from their moral strictness, or piety, if that term is preferred. Despite their extreme courage, the Americans found them unsatisfactory allies in their operations against the Apaches.' [Ibid., 182-4.] However much the details and variations of the ceremonies of expiation and purification after the slaying of enemies might be

II. TABOO AND EMOTIONAL AMBIVALENCE

41

of interest for deeper research into the subject, I shall break off at this point, since for our present purpose they have nothing more to tell us. I may perhaps suggest that the temporary or permanent isolation of professional executioners, which has per­ sisted to the present day, may belong in this connection. The position of the public hangman in mediaeval society offers a good picture of the workings of taboo among savages. 1 In the accepted explanation of all these observances of appease­ ment, restriction, expiation and purification, two principles are combined: the extension of the taboo from the slain man on to everything that 'has come in contact with him, and the fear of the slain man's ghost. How these two factors are to be com­ bined with each other to explain the ceremonials, whether they are to be regarded as of equal weight, whether one is primary and the other secondary, and if so which-none of these ques­ tions receives an answer, and indeed it would be hard to find one. We, on the other hand, can lay stress on the uniry of our view, which derives all of these observances from emotional ambivalence towards the enemy. (b) The Taboo upon Rulers

The attitude of primitive peoples to their chiefs, kings and priests is governed by two basic principles which seem to be complementary rather than contradictory. A ruler 'must not only be guarded, he must also be guarded against'. (Frazer, 19IIb, 132.) Both of these ends are secured by innumerable taboo observances. We know already why it is that rulers must be guarded against. It is because they are vehicles of the mys­ terious and dangerous magical power which is transmitted by contact like an electric charge and which brings death and ruin to anyone who is not protected by a similar charge. Any im­ mediate or indirect contact with this dangerous sacred entity is therefore avoided; and, if it cannot be avoided, some ceremon­ ial is devised to avert the dreaded consequences. The Nubas of East Africa, for instance, · 'believe that they would die if they 1 Further examples of these practices will be ound in Frazer ( 1 g I I b, 165-90) in the section upon 'Manslayers tabooed'.

42

TOTEM AND TABOO

entered the house of their priestly king; however they can evade the penalty of their intrusion by baring the left shoulder and getting the king to lay his hand on it'. [Loe. cit.] Here we are met by the remarkable fact that contact with the king is a remedy and protection against the dangers provoked by con­ tact with the king. No doubt, however, there is a contrast to be drawn between the remedial power of a touch made deliber­ ately by the king and the danger which arises if he is touched­ a contrast between a passive and an active relation to the king. For examples of the healing power of the royal touch there is no need to resort to savages. The kings of England, in times that are not yet remote, enjoyed the power of curing scrofula, which was known accordingly as 'th� King's Evil'. Queen Elizabeth I exercised this royal prerogative no less than her successors. Charles I is said to have cured a hundred patients at a stroke in 1633. But it was after the Restoration of the monarchy under his dissolute son, Charles II, that the royal cures of scrofula reached their climax. In the course of his reign he is reputed to have touched close upon a hundred thousand persons. The crowd of those in search of cure used to be so great that on one occasion six or seven of those who came to be healed were trampled to death. The sceptical William of Orange, who be­ came King of England after the dismissal of the Stuarts, refused to lend himself to these magical practices. On the only occasion on which he was persuaded into laying his hands on a patient, he said to him: 'God give you better health and more sense.' '(Frazer, 1911a, 1, 368-70.) The stories which follow are evidence of the fearful effects of active contact made, even unintentionally, with a king or any­ thing belonging to him. 'It once happened that a New Zealand chief of high rank and great sanctity had left the remains of his dinner by the wayside. A slave, a stout, hungry fellow, coming up after the chief had gone, saw the unfinished dinner, and ate it up without asking questions. Hardly had he finished when he was informed by a horror-stricken spectator that the food of which he had eaten was the chief's.' He was a strong, courage­ ous man, but 'no sooner did he hear the fatal news than he was seized with the most extraordinary convulsions and cramp in the stomach, which never ceased till he died, about sundown

•

11. TABOO AND EMOTIONAL AMBIVALENCE

43

the same day'. 'A Maori woman having eaten of some fruit, and being afterwards told that the fruit had been taken from a tabooed place, exclaimed that the spirit of the chief, whose sanctity had been thus profaned, would kill her. This was in the afternoon, and next day by twelve o'clock she was dead.'• 'A Maori chief's tinder-box was once the means of killing several persons; for, having been lost by him, and found by some men who used it to light their pipes, they died of fright on learning to whom it had belonged.'• It is not to be wondered at that a need was felt for isolating such dangerous persons as chiefs and priests from the rest of the fommunity-to build a barrier round them which would make ,them inaccessible. It may begin to dawn on us that this barrier, originally erected for the observance of taboo, exists to this day in the form of court ceremonial. But perhaps the major part of this taboo upon rulers is not derived from the need for protection against them. The second reason for the special treatment of privileged persons-the need to provide protectionfor them against the threat of danger-has had an obvious part in creating taboos and so of giving rise to court etiquette. The need to protect the king from every possible form of danger follows from his immense importance to his subjects, whether for weal or woe. It is his person which, strictly speaking, regulates the whole course of existence. 'The people have to thank him for the rain and sunshine which foster the fruits of the earth, for the wind which brings ships to their coasts, and even for the solid ground beneath their feet.' (Frazer, 1911b, 7.) These rulers among savage peoples possess a degree of power and a capacity to confer benefits which are an attribute only of gods, and with which at later stages of civilization only the most servile of courtiers would pretend to credit them. 1

It must strike us as self-contradictory that persons of such un­ limited power should need to be protected so carefully from the threat of danger; but that is not the only contradiction shown in 1

Frazer (1911b, 134-5), quoting a Pakeha Maori (1884 [96 f.]). Frazer (loc. cit.), quoting Brown (1845 [76]). 8 Frazer (loc. cit.) [quoting Taylor (1870, 164)].

2

44

TOTEM AND TABOO

the treatment of royal personages among savage peoples. For these peoples also think it necessary to keep a watch on their king to see that he makes a proper use of his powers; they feel by no means convinced of his good intentions or conscientiousness. Thus an element of distrust may be traced among the reasons for the taboo observances that surround the king. 'The idea', writes Frazer (1911b, 7 f.), 'that early kingdoms are despotisms in which the people exist only for the sovereign, is wholly in­ applicable to the monarchies we are considering. On the con­ trary, the sovereign in them exists only for his subjects; his life is only valuable so long as he discharges the duties of his position by ordering the course of nature for his people's benefit. So soon as he fails to do so, the care, the devotion, the religious homage which they had hitherto lavished on him cease and are changed into hatred and contempt; he is dismissed ignominiously, and may be thankful if he escapes with his life. Worshipped as a god one day, he is killed as a criminal the next. But in this changed behaviour of the people there is nothing capricious or incon­ stant. On the contrary, their conduct is entirely of a piece. If their king is their god, he is or should be also their preserver; and if he will not preserve them, he must make room for another who will. So long, however, as he answers their expectations, there is no limit to the care which they take of him, and which they compel him to take of himself. A king of this sort lives hedged in by a ceremonious etiquette, a network of prohibitions and observances, of which the intention is not to contribute to his dignity, much less to his comfort, but to restrain him from conduct which, by disturbing the harmony of nature, might involve himself, his people, and the universe in one common catastrophe. Far from adding to his comfort, these observances, by trammelling his every act, annihilate his freedom and often render the very life, which it is their object to preserve, a burden and sorrow to him.' One of the most glaring instances of a sacred ruler being fet­ tered and paralysed in this way by taboo ceremonials is to be found in the mode of life of the Mikado of Japan in earlier cen­ turies. An account written more than two hundred years ago reports that the Mikado 'thinks it would be very prejudicial to his dignity and holiness to touch the ground with his feet; for

II. TABOO AND EMOTIONAL AMBIVALENCE

45

this reason, when he intends to go anywhere, he must be carried thither on men's shoulders. Much less will they suffer that he should expose his sacred person to the open air, and the sun is not thought worthy to shine on his head. There is such a holi­ ness ascribed to all parts of his body that he dares to cut off neither his hair, nor his beard, nor his nails. However, lest he should grow too dirty, they may clean him in the night when he is asleep; because, they say, that which is taken from his body at that time hath been stolen from him and that such a theft doth not prejudice his holiness or dignity. In ancient times he was obliged to sit on the throne for some hours every morning, with the imperial crown on his head, but to sit altogether like a statue, without stirring either hands or feet, head or eyes, nor indeed any part of his body, because, by this means, it was thought that he could preserve peace and tranquillity in his empire; for if, unfortunately, he turned himself on one side or the other, or if he looked a good while towards any part of his dominions, it was apprehended that war, famine, fire, or some other great misfortune was near at hand to desolate the country.' 1 Some of the taboos laid upon barbarian kings remind one vividly of the restrictions imposed upon murderers. Thus in West Africa, 'at Shark Point near Cape Padron, in Lower Guinea, lives the priestly king Kukulu, alone in a wood. He may not touch a woman nor leave his house; indeed he may not even quit his chair, in which he is obliged to sleep sitting, for if he lay down no wind would arise and navigation would be stopped. He regulates storms, and in general maintains a wholesome and equable state of the atmosphere.' The same writer says of Loango {in the same part of the world) that the more powerful a king is, the more taboos he is bound to observe.• The heir to the throne is also subject to them from infancy; their number increases as he advances in life, till at the moment that he ascends the throne he is positively suffocated by them. Our space will not allow nor does our interest require us to enter further into a description of the taboos associated with the dignity of kings and priests. I will only add that the principal 1

1

Kaempfer (1727 [1, 150]), quoted by Frazer (1911b, 3 f.). Frazer (1911b, 5 and 8), quoting Bastian (1874-5 [1, 287 and 355]).

S.F. XIII-E

46

TOTEM AND TABOO

part is played in them by restrictions upon freedom of move­ ment and upon diet. Two examples oftaboo ceremonials occur­ ring in civilized communities ofa far higher level ofculture will serve to show, however, what a conservative effect upon ancient usages is exercised by contact with these privileged personages. The Flamen Dialis, the high priest ofJupiter in ancient Rome, was obliged to observe an extraordinary number of taboos. He 'might not ride or even touch a horse, nor see an army under arms, nor wear a ring which was not broken, nor have a knot on any part of his garments; ... he might not touch wheaten flour or leavened bread; he might not touch or even name a goat, a dog, raw meat, beans, and ivy; ... his hair could be cut only by a free man and with a bronze knife, and his hair and nails when cut had to be buried under a lucky tree; ... he might not touch a dead body; ... he might not be uncovered in the open air', and so on. 'His wife, the Flaminica, had to observe nearly the same rules, and others ofher own besides. She might not ascend more than three steps of the kind of staircase called Greek; at a certain festival she might not comb her hair; the leather of he1 shoes might not be made from a beast that had died a natural death, but only from one that had been slain or sacrificed; ifshe heard thunder she was tabooed till she had offered an expiatory sacrifice.' (Frazer, 1911b, 13 f.) The ancient kings of Ireland were subject to a number ofex­ ceedingly strange restrictions. If these were obeyed, every kind of blessing would descend upon the country, but if they were violated, disasters ofevery kind would visit it. A complete list of these taboos is contained in the Book of Rights, the two oldest manuscript copies of which date from 1390 and 1418. The pro­ hibitions are ofthe most detailed character, and refer to specific actions at specific places at specific times: the king, for instance, may not stay in a certain town on a particular day of the week; he may not cross a certain river at a particular hour ofthe day; he may not encamp for nine days on a certain plain, and so on. (Frazer, 19ub, II f.) Among many savage peoples the severity of these taboo re• strictions upon priestly kings has led to consequences which have been important historically and are of particular interest fromour point ofview. The dignity oftheir position ceased to be

II. TABOO AND EMOTIONAL AMBIVALENCE

17

an enviable thing, and those who were offored it often took every possible mi;ans of escaping it. Thus in Cambodia, where there are kingships of Fire and Water, it is often necessary to force successors into accepting these distinctions. On Niue or Savage Island, a coral island in the South Pacific, the monarchy actually came to an end because no one could be induced to take over the responsible and dangerous office. 'In some parts of West Africa, when the king dies, a family council is secretly held to determine his successor. He on whom the choice falls is sud­ denly seized, bound, and thrown into the fetish-house, where he is kept in durance till he consents to accept the crown. Some­ times the heir finds means of evading the honour which it is thought to thrust upon him; a ferocious chief has been known to go about constantly armed, resolute to resist by force any attempt to set him on the throne.' 1 Among the natives of Sierra Leone the objection to accepting the honour of kingship became so great that most tribes were obliged to choose foreigners as their kings. Frazer (19IIb, 17-25) attributes to these circumstances the fact that in the course of history there eventually came about a division of the original priestly kingship into a spiritual and a temporal power. Weighed down by the burden of their sacred office, kings became unable to exert their dominance in real affairs and these were left in the hands of inferior but practical persons, who were ready to renounce the honours of kingship. These, then, became the temporal rulers, while spiritual suprem­ acy, deprived of any practical significance, was left to the former taboo )ungs. It is familiar knowledge how far this hypothesis finds confirmation in the history of old Japan. If we take a general survey of the relations of primitive men to their rulers, we are left with an expectation that we shall have no great difficulty in advancing from a description of them to a psycho-analytic understanding of them. Those relations are of a complex kind and not free from contradictions. Rulers are allowed great privileges, which coincide exactly with the taboo prohibitions imposed on other people. They are privileged per­ sons: they may do or enjoy precisely what other people are for1

Frazer (1911b, 17 f.-), quoting Bastian (1874-5 [1, 354 and 2, g]).

50

TOTEM AND TABOO

Another side of the attitude of primitive peoples towards their rulers recalls a procedure which is common in neuroses gener­ ally but comes into the open in what are known as delusions of persecution. The importance of one particular person is im­ mensely exaggerated and his absolute power is magnified to the most improbable degree, in order that it may be easier to make him responsible for everything disagreeable that the patient may experience. Savages are really behaving in just the same way with their kings when they ascribe to them power over rain and sunshine, wind and weather, and then depose them or kill them because Nature disappoints their hopes of a successful hunt or a rich harvest. The model upon which paranoics base their delusions of persecution is the relation of a child to his father. A son's picture of his father is habitually clothed with excessive powers of this kind, and it is found that distrust of the father is intimately linked with admiration for him. When a paranoic turns the figure of one of his associates into a 'perse­ cutor', he is raising him to the rank of a father: he is putti11g him into a position in which he can blame him for all his misfor­ tunes. Thus this second analogy between savages and neurotics gives us a glimpse of the truth that much of a savage's attitude to his ruler is derived from a child's infantile attitude to his father. But the strongest support for our effort to equate taboo pro­ hibitions with neurotic symptoms is to be found in the taboo ceremonials themselves, the effect of which upon the position of , royalty has already been discussed. These ceremonials unmis­ takably reveal their double meaning and their derivation from ambivalent impulses, as soon as we are ready to allow that the results which they bring about were intended from the first. The taboo does not only pick out the king and exalt him above all common mortals, it also makes his existence a torment and an intolerable burden and reduces him to a bondage far worse than that of his subjects. Here, then, we have an exact counterpart of the obsessional act in the neurosis, in which the suppressed im­ pulse and the impulse that suppresses it find simultaneous and common satisfaction. The obsessional act is ostensibly a protection against the prohibited act; but actually, in our view, it is a repeti­ tion of it. The 'ostensibly' applies to the conscious part of the

II. TABOO AND EMOTIONAL AMBIVALENCE

51

mind, and the 'actually' to the unconscious part. In exactly the same way, the ceremonial tabo,o of kings is ostensibly the highest honour and protection for them, while actually it is a punishment for their exaltation, a revenge taken on them by their subjects. The experiences of Sancho Panza (as described by Cervantes) when he was Governor of his island convinced him that this view of court ceremonial was the only one that met the case. If we could hear the views of modern kings and rulers on the sub­ ject, we might find that there were many others who agreed with him. The question of why the emotional attitude towards rulers includes such a powerful unconscious element of hostility raises a very interesting problem, but one that lies outside the limits of the present study. I have already hinted at the fact that the child's complex of emotions towards his father-the father-com­ plex-has a bearing on the subject, and I may add that more information on the early history of the kingship would throw a decisive light on it. Frazer (xgua) has put forward impressive reasons, though, as he himself admits, not wholly conclusive ones, for supposing that the earliest kings were foreigners who, after a brief reign, were sacrificed with solemn festivities as representatives of the deity. It is possible that the course taken by the evolution of kings :may also liave had an influence upon the myths of Christendom.

(c) The Taboo upon the Dead We kn.ow that the dead are powerful rulers; but we may per­ haps be surprised when we learn that they are treated as enemies. The taboo upon the dead is-if I may revert to the simile of infection-especially virulent among most primitive peoples. It is manifested, in the first instance, in the consequences that fol­ low contact with the dead and in the treatment of mourners. Among the Maoris anyone who had handled a corpse or taken any part in its burial was in the highest degree unclean and was almost cut off from intercourse with his fellow-men, or, as we might put it, was boycotted. He could not enter any house, or come into co1:tact with any person or thing without infecting them. He might not even touch food with his hands,

52

I I

TOTEM AND TABOO

which, owing to their uncleanness, had become quite useless. 'Food would be set for him on the ground, and he would then sit or kneel down, and, with his hands carefully held behind his back, would gnaw at it as best he could. In some cases he would be fed by another person, who with outstretched arm contrived to do it without touching the tabooed man; but the feeder was himself subjected to many severe restrictions, little less onerous than those which were imposed upon the other. In almost every populous village there lived a degraded wretch, the low­ est of the low, who earned a sorry pittance by thus waiting upon the defiled.' He alone was allowed 'to associate at arm's length with one who had paid the last offices . . . to the dead. And when, the dismal term of his seclusion being over, the mourner was about to mix with his fellows once more, all the dishes he had used in his seclusion were diligently smashed, and all the garments he had worn were carefully thrown away.' [Frazer, 1911b, 138 f.] The taboo observances after bodily contact with the dead are the same over the whole of Polynesia, Melanesia and a part of Africa. Their most regular feature is the prohibition against those who have had such contact touching food themselves, and the consequent necessity for their being fed by other people. It is a remarkable fact that in Polynesia (though the report may perhaps refer only to Hawaii) priestly kings were subject to the same restriction while performing their sacred functions. 1 The case of the taboo upon the dead in Tonga offers a specially clear instance of the way in which the degree of prohibition varies according to the taboo power of the person upon whom the taboo is imposed. Thus anyone who touches a dead chief is un­ clean for ten months; but if he himself is a chief he is only tabooed for three, four, or five months according to the rank of the dead man; but if the dead man were the 'great divine chief', even the greatest chief would be tabooed for ten months. These savages believe firmly that anyone who violates the taboo ordinances is bound to fall ill and die; indeed they believe it so firmly that, in the opinion of an observer, 'no native ever made an experiment to prove the contrary'.1 1

1

Frazer (loc. cit.) [quoting Ellis (1832-6, 4, 388)]. Frazer (1911b, 140), quoting Mariner (1818 [1, 141]).

II. TABOO AND EMOTIONAL AMBIVALENCE

53

Essentially the same prohibitions (though from our point of view they are more interesting) apply to those who have been in contact with the dead only in a metaphorical sense: the dead person's mourning relations, widowers and widows.The obser­ vances that we have so far mentioned may seem merely to give characteristic expression to the virulence of the taboo and its contagious power.But those which now follow give us a hint at the reasons for the taboo-both the ostensible ones and what we must regard as the deep-lying real ones. 'Among the Shuswap of British Columbia widows and widowers in mourning are secluded and forbidden to touch their own head or body; the cups and cooking vessels which they use may be used by no one else .... No hunter would come near such mourners, for their presence is unlucky. If their shadow were to fall on anyone, he would be taken ill at once.They em­ ploy thorn-bushes for bed and pillow ... and thorn-bushes are also laid all around their beds.' 1 This last measure is designed to keep the dead person's ghost at a distance.The same purpose is shown still more clearly in the usage reported from another North American tribe which provides that, after her husband's death, 'a widow would wear a breech-cloth made of dry bunch­ grass for several days to prevent her husband's ghost having intercourse with her.'1 This suggests that contact 'in a meta­ phorical sense' is after all understood as being bodily contact, for the dead man's ghost does not leave his relations and does not cease to 'hover' round them during the time of mourning. 'Among the Agutainos, who inhabit Palawan, one of the Philippine Islands, a widow may not leave her hut for seven or eight days after the death; and even then she may only go out at an hour when she is not likely to meet anybody, for whoever looks upon her dies a sudden death. To prevent this fatal catas­ trophe, the widow knocks with a wooden peg on the trees as she goes along, thus warning people of her dangerous proximity; and the very trees on which she knocks soon die.'8 The nature of the danger feared from a widow such as this is made plain by another example. 'In the Mekeo district of British New Guinea [Frazer (1911b, 142), quoting Boas (1890, 643 f.).] [Frazer (1911b, 143), quoting Teit (1900, 332 f.).] 1 [Frazer (1911b, 144), quoting Blumentritt (1891, 182).] 1

2

54

TOTEM AND TABOO

a widower loses all his civil rights and becomes a social outcast, an object offear and horror, shunned by all. He may not culti­ vate a garden, nor show himself in public, nor walk on the roads and paths. Like a wild beast he must skulk in the long grass and the bushes; and if he sees or hears anyone coming, especially a woman, he must hide behind a tree or a thicket.' 1 This last hint makes it easy to trace the origin of the dangerous character of widowers or widows to the danger of temptation. A man who has lost his wife must resist a desire to find a substitute for her; a widow must fight against the same wish and is more­ over liable, being without a lord and master, to arouse the de­ sires of other men. Substitutive satisfactions of such a kind run counter to the sense of mourning and they would inevitably kindle the ghost's wrath.• One of the most puzzling, but at the same time instructive, usages in connection with mourning is the prohibition against uttering the name of the dead person. This custom is extremely widespread, it is expressed in a variety of ways and has had im­ portant consequences. It is found not only among the Austra­ lians and Polynesians (who usually show us taboo observances in the best state of preservation), but also among 'peoples so widely separated from each other as the Samoyeds of Siberia and the Todas of southern India; the Mongols of Tartary and the Tuaregs of the Sahara; the Ainos of Japan and the Akamba and Nandi of central Africa; the Tinguianes of the Philippines and the inhabitants of the Nicobar Islands, of Borneo, of Mada­ gascar, and of Tasmania.' (Frazer, 1911b, 353.) In some of these cases the prohibition and its consequences last only during the period of mourning, in others they are permanent; but it seemsinvariably to diminish in strictness with the passage of time. The avoidance of the name of a dead person is as a rule en­ forced with extreme severity. Thus in some South American tribes it is regarded as a deadly insult to the survivors to mention 1

[Frazer (1911b, 144), quoting Guis (1902, 208 f.).] The patient whose 'impossibilities' I compared with taboos earlier in this paper (see page 28) told me that whenever she met anyone dressed in mourning in the street she was filled with indignation: such people, she thought, should be forbidden to go out. 1

II. TABOO AND EMOTIONAL AMBIVALENCE

55

the name of a dead relative in their presence, and the punish­ ment for it is not less than that laid down for murder. (Ibid., 352.) It is not easy at first to see why the mention of the name should be regarded with such horror; but the dangers involved have given rise to a whole number of methods of evasion which are interesting and important in various ways. Thus the Masai in East Africa resort to the device of changing the dead man's name immediately after his death; he may then be mentioned freely under his new name while all the restrictions remain attached to the old one. This seems to presuppose that the dead man's ghost does not know and will not get to know his new name. [Ibid., 354.] The Adelaide and Encounter Bay tribes of South Australia are so consistently careful that after a death everyone bearing the same name as the dead man's, or a very similar one, changes it for another. [Ibid., 355.] In some in­ stances, as for instance among certain tribes in Victoria and in North-West America, this is carried a step further, and after a death all the dead person's relations change their names, irre­ spective of any similarity in their sound. [Ibid., 357.] Indeed, among the Guaycurus in Paraguay, when a death had taken place, the chief used to change the name of every member of the tribe; and 'from that moment everybody remembered his new name just as if he had borne it all his life'. 1 Moreover, if the name of the dead man happens to be the same as that of an animal or common object, some tribes think it necessary to give these animals or objects new names, so that the use of the former names shall not recall the dead man to memory. This usage leads to a perpetual change of vocabulary, which causes much difficulty to the missionaries, especially when such changes are permanent. In the seven years which the missionary Dobrizhoffer spent among the Abipones of Para­ guay, 'the native word for jaguar was changed thrice, and the words for crocodile, thorn, and the slaughter of cattle under­ went similar though less varied vicissitudes'.1 The dread of uttering a dead person's name extends, indeed, to an avoidance of the mention of anything in which the dead man played a 1 Frazer (1911b, 357), quoting an old Spanish observer [Lozano (1733, 70)]. 1 Frazer (1911b,360), quoting Dobrizhoffer [1784, 2,301].

I

'l

56

TOTEM AND TABOO

part; and an important consequence of this process of suppres­ sion is that these peoples possess no tradition and no historical memory, so that any research into their early history is faced by the greatest difficulties. [Ibid., 362 f.] A number of these primi­ tive races have, however, adopted compensatory usages which revive the names of dead persons after a long period of mourn­ ing by giving them to children, who are thus regarded as rein­ carnations of the dead. [Ibid., 364 f.] This taboo upon names will seem less puzzling if we bear in mind the fact that savages regard a name as an essential part of a man's personality and as an important possession: they treat words in every sense as things. As I have pointed out elsewhere [Freud, 1905c, Chap. IV], our own children do the same. They are never ready to accept a similarity between two words as having no meaning; they consistently assume that if two things are called by similar-sounding names this must imply the existence of some deep-lying point of agreement between them. Even a civilized adult may be able to infer from certain peculi­ arities in his own behaviour that he is not so far removed as he may have thought from attributing importance to proper names, and that his own name has become to a very remarkable extent bound up with his personality. So, too, psycho-analytic practice comes upon frequent confirmations of this in the evi­ dence it finds of the importance of names in unconscious mental activities. 1 As was only to be expected, obsessional neurotics behave exactly like savages in relation to names. Like other neurotics, they show a high degree of 'complexive sensitiveness'• in regard to uttering or hearing particular words and names; and their attitude towards their own names imposes numerous, and often serious, inhibitions upon them. One of these taboo patients of my acquaintance had adopted a rule against writing her own name, for fear that it might fall into the hands of someone who would then be in possession of a portion of her personality. She was obliged to fight with convulsive loyalty against the tempta­ tions to which her imagination subjected her, and so forbade 1 Cf.

Stekel [1911] and Abraham [1911]. ['Komplexemp.findlichkeit'-a term used by Jung in connection with his word-association experiments.] 1

,,

II. TABOO AND EMOTIONAL AMBIVALENCE

57

herself 'to surrender any part of her person'. This included in the first place her name, and later extended to her handwriting, till finally she gave up writing altogether. We shall no longer feel surprised, therefore, at savages regard­ ing the name of a dead person as a portion of his personality and making it subject to the relevant taboo. So, too, uttering the name of a dead person is clearly a derivative of having contact with him. We may therefore turn to the wider problem of why such contact is submitted to so strict a taboo. The most obvious explanation would point to horror roused by dead bodies and by the. changes which quickly become vis­ ible in them. Some part must also be played in the matter by mourning for the dead person, since it must be a motive force in everything relating to him. But horror at the corpse clearly does not account for all the details of the taboo observances, and mourning cannot explain why the uttering of the dead man's name is an insult to his survivors. Mourning, on the contrary, tends to be preoccupied with the dead man, to dwell upon his memory and to preserve it as long as possible. Something other than mourning must be held responsible for the peculiarities of the taboo usages, something which has very different purposes in view. It is precisely the taboo upon names that gives us the clue to this unknown motive; and if the usages alone did not tell us, we should learn it from what the mourning savages say to us themselves. For they make no disguise of the fact that they are afraid of the presence or of the return of the dead person's ghost; and they perform a great number of ceremonies to keep him at a distance or drive him off. 1 They feel that to utter his name is equivalent to invoking him and will quickly be followed by his presence. 1 And ?.ccordingly they do everything they can to avoid any such evocation. They disguise themselves so that the ghost shall not recognize them, 8 or they change his name or 1 Frazer (1911b, 353) mentions the Tuaregs of the Sahara as an example of this explanation being given by the savages themselves. 1 Subject, perhaps, to the condition that some of his bodily remains are still in existence. (Ibid., 372.) 8 In the Nicobar Islands. (Ibid., 358.)

'

58

,.

I' II

:I I I

.!

I, I

TOTEM AND TABOO

their own; they are furious with reckless strangers who by utter­ ing the ghost's name incite him against the survivors. It is im­ possible to escape the conclusion that, in the words of Wundt (1906, 49), they are victims to a fear of 'the dead man's soul which has become a demon'. Here, then, we seem to have found a confirmation of Wundt's view, which, as we have al­ ready seen (p. 24), considers that the essence of taboo is a fear of demons. This theory is based on a supposition so extraordinary that it seems at first sight incredible: the supposition, namely, that a dearly loved relative at the moment of his death changes into a demon, from whom his survivors can expect nothing but hos­ tility and against whose evil desires they must protect them­ selves by every possible means. Nevertheless, almost all the authorities are at one in attributing these views to primitive peoples. Westermarck, who, in my opinion, takes far too little notice of taboo in his book on The Origin and Development of the Moral Ideas, actually writes in his chapter on 'Regard for the Dead': 'Generally speaking, my collection of facts has led me to the conclusion that the dead are more commonly regarded as enemies than friends, and that Professor Jevons and Mr. Grant Allen are mistaken in their assertion that, according to early beliefs, the malevolence of the dead is for the most part directed against strangers only, whereas they exercise a fatherly care over the lives and fortunes of their descendants and fellow clansmen.' 1 In an interesting volume, Rudolf Kleinpaul (1898) has used the i:-emnants among civilized races of the ancient belief in 1 Westermarck (1906-8, 2, 532 ff.). In his footnotes and in the section of the text which follows, the author gives copious confirmatory evidence, often of an extremely pertinent sort. For instance: 'Among the Maoris the nearest and most beloved relatives were supposed to have their natures changed by death, and to become malignant, even towards those they formerly loved. [Quoting Taylor (1870, 18).] .•. Australian natives believed that a deceased person is malevolent for a long time after death, and the more nearly related the more he is feared. [Quoting Fraser (1892, 80).] ... According to ideas prevalent among the Central Eskimo, the dead are at first malevolent spirits who frequently roam around the villages, causing sickness and mischief and killing men by their touch; but subsequently they are supposed to attain rest and are no longer feared. [Quoting Boas (1888, 591).]'

II. TABOO AND EMOTIONAL AMBIVALENCE

59

spirits to throw light on the relation between the living and the dead. He, too, reaches the final conclusion that the dead, filled with a lust for murder, sought to drag the living in the�. train. The dead slew; and the skeleton which we use to-day to picture the dead stands for the fact that they themselves were slayers. The living did not feel safe from the attacks of the dead till there was a sheet of water between them. That is why men liked to bury the dead on islands or on the farther side of rivers; and that, in turn, is the origin of such phrases as 'Here and in the Beyond . ' Later, the malignity of the dead diminished and was restricted to special categories which had a particular right to feel resent­ ment-such as murdered men, for instance, who in the form of evil spirits went in pursuit of their murderers, or brides who had died with their desires unsatisfied. But originally, says Kleinpaul, all of the dead were vampires, all of them had a grudge against the living and sought to injure them and rob them of their lives. It wa� from corpses that the concept of evil spirits first arose. The hypothesis that after their death those most beloved were transformed into demons clearly raises further questions. What was it that induced primitive men to attribute such a change of feeling to those who had been dear to them? Why did they make them into demons? Westermarck (1906-8, 2, 534 f.) is of the opinion that these questions can be answered easily. 'Death is commonly regarcled as the gravest of all misfortunes; hence the dead are believed to be exceedingly dissatisfied with their fate. According to primitive ideas a person only dies if he is killed­ by magic if not by force-and such a death naturally tends to make the soul revengeful and ill-tempered. It is envious of the living and is longing for the company of its old friends; no won­ der, then, that it sends them diseases to cause their death .... But the notion that the disembodied soul is on the whole a mali­ cious being ...is also, no doubt, intimately connected with the instinctive fear of the dead, which is in its turn the outcome of the fear of death.' The study of psychoneurotic disorders suggests a more com­ prehensive explanation, which at the same time covers that put forward by Westermarck.

60

TOTEM AND TABOO

When a wife has lost lier husband or a daughter her mother, it not infrequently happens that the survivor is overwhelmed by tormenting doubts (to which we give the name of'obsessive self­ reproaches') as to whether she may not herself have been responsible for the death of this cherished being through some act of carelessness or neglect. No amount of recollection of the care she lavished on the sufferer, no amount of objective dis­ proof of the accusation, serves to bring the torment to an end. It may be regarded as a pathological form of mourning, and with the passage of time it gradually dies away. The psycho­ analytic investigation of such cases has revealed the secret motives of the disorder. We find that in a certain sense these ob­ sessive self-reproaches are justified, and that this is why they are proof against contradictions and protests. It is not that the mourner was really responsible for the death or was really guilty of neglect, as the self-reproaches declare to be the case. None the less there was something in her-a wish that was un­ conscious to herself-which would not have been dissatisfied by the occurrence of death and which might actually have brought it about if it had had the power. And after death has occurred, it is against this unconscious wish that the reproaches are a reaction. In almost every case where there is an intense emo­ tional attachment to a particular person we find that behind the tender love there is a concealed hostility in the unconscious. This is the classical example, the prototype, of the ambivalence of human emotions. This ambivalence is present to a greater or less amount in the innate disposition of everyone; normally, there is not so much of it as to produce the ·obsessive self­ reproaches we are considering. Where, however, it is copiously present in the disposition, it will manifest itself precisely in the subject's relation to those of whom he is most fond, in the place, in fact, where one would least expect to find it. It must be sup­ posed that the presence of a particularly large amount of this original emotional ambivalence is characteristic of the disposi­ tion of obsessional neurotics-whom I have so often brought up for comparison in this discussion upon taboo. We have now discovered a motive which can explain the idea that the souls of those who have just died are transformed into demons and the necessity felt by survivors to protect thems�lves

II. TABOO AND EMOTIONAL AMBIVALENCE

61

by taboos against their hostility. Let us suppose that the emo­ tional life of primitive peoples is characterized by an amount of ambivalence as great as that which we are led by the findings of psycho-analysis to attribute to obsessional patients. It then becomes easy to understand how after a painful bereavement savages should be obliged to produce a reaction against the hostility latent in their unconscious similar to that expressed as obsessive self-reproach in the case of neurotics. But this hostility, distressingly felt in the unconscious as satisfaction over the death, is differently dealt with among primitive peoples. The defence against it takes the form of displacing it on to the object of the hostility, on to the dead themselves. This defensive pro­ cedure, which is a common one both in normal and in patho­ logical mental life, is known as a 'projection'. The survivor thus denies that he has ever harboured any hostile feelings against the dead loved one; the soul of the dead harbours them instead and seeks to put them into action during the whole period of mourning. In spite of the successful defence which the survivor achieves by means of projection, his emotional reaction shows the characteristics of punishment and remorse, for he is the sub­ ject of fears and submits to renunciations and restrictions, though these are in part disguised as measures of protection against the hostile demon. Once again, therefore, we find that the taboo has grown up on the basis of an ambivalent emotional attitude. The taboo upon'the dead arises, like the others, from the contrast between conscious pain and unconscious satisfac­ tion over the death that has occurred. Since such is the origin of the ghost's resentment, it follows naturally that the survivors who have the most to fear will be those who were formerly its nearest and dearest. In this respect taboo observances, like neurotic symptoms, have a double sense. On the one hand, in their restrictive char­ acter, they are expressions of mourning; but on the other hand they clearly betray-what they seek to conceal-hostility against the dead disguised as self-defence. We have already learned that certain taboos arise out of fear of temptation. The fact that a dead man is helpless is bound to act as an encourage­ ment to the survivor to give free rein to his hostile passions, and that te�ptation must be countered by a prohibition. S.I:''. Xlll-J,'

62

I'

'' ,I .,

r I

• I

TOTEM AND TABOO

Westermarck is right in insisting that savages draw no dis­ tinction between violent and natural death. In the view of un­ conscious thinking, a man who has died a natural death is a murdered man: evil wishes have killed him. 1 Anyone who in­ vestigates the origin and significance of dreams of the death of loved relatives (of parents or brothers or sisters) will be able to convince himself that dreamers, children and savages are at one in their attitude towards the dead-an attitude based upon emotional ambivalence. 8 At the beginning of this essay [p. 24] disagreement was ex­ pressed with Wundt's opinion that the essence of taboo was a fear of demons. Yet we have now assented to an explanation that derives the taboo upon the dead from a fear of the soul of the dead person transformed into a demon. The apparent contradiction can easily be resolved. It is true that we have accepted the presence of demons, but not as something ultimate and psychologically unanalysable. We have succeeded, as it were, in getting behind the demons, for we have explained them as projections of hostile feelings harboured by the survivors against the dead. Both of the two sets offeelings (the affectionate and the hos­ tile), which, as we have good reason to believe, exist towards the dead person, seek to take effect at the time of the bereave­ ment, as mourning and as satisfaction. There is bound to be a conflict between these two contrary feelings; and, since one of the two, the hostility, is wholly or for the greater part uncon­ scious, the outcome of the conflict cannot be to subtract, as it were, the feeling with the lesser intensity from that with the greater and to establish the remainder in consciousness-as occurs, for instance, when one forgives a slight that one has re­ ceived from someone of whom one is fond. The process is dealt with instead by the special psychical mechanism known in psycho-analysis, as I have said, by the name of'projection'. The hostility, of which the survivors know nothing and moreover wish to know nothing, is ejected from internal perception into the external world, and thus detached from them and pushed 1

Cf. the next essay in this volume. [Cf. Freud, The Interpretation of Dreams 248 ff.] 2

(1900a), Standard Ed.,

4,

... II. TABOO AND EMOTIONAL AMBIVALENCE

63

on to someone else. It is no longer true that they are rejoicing to be rid of the dead man; on the contrary, they are mourning for him; but, strange to say, he has turned into a wicked demon ready to gloat over their misfortunes and eager to kill them. It then becomes necessary for them, the survivors, to defend them­ selves against this evil enemy; they are relieved of pressure from within, but have only exchanged it for oppression from without. It cannot be disputed that this process of projection, which turns a dead man into a malignant enemy, is able to find sup­ port in any real acts of hostility on his part that may be recol­ lected and felt as a grudge against him: his severity, his love of power, his unfairness, or whatever else may form the back­ ground of even the tenderest of human relationships. But it can­ not be such a simple matter as that. This factor alone cannot explain the creation of demons by projection. The faults of the dead no doubt provide a part of the explanation of the survivors' hostility; but they would not operate in this way unless the sur,­ vivors had first developed hostility on their own account. The moment of death, moreover, would certainly seem to be a most inappropriate occasion for recalling any justifiable grounds of complaint that might exist. It is impossible to escape the fact that the true determining factor is invariably unconscious hos­ tility. A hostile current of feeling such as this against a person's nearest and dearest relatives may remain latent during their lifetime, that is, its existence may not be betrayed to conscious­ ness either directly or through some substitute. But when they die this is no longer possible and the conflict becomes acute. The mourning which derives from an intensification of the affectionate feelings becomes on the one hand more impatient of the latent hostility and, on the other hand, will not allow it to give rise to any sense of satisfaction. Accordingly, there follow the repression of the unconscious hostility by the method of pro­ jection and the construction of the ceremonial which gives ex­ pression to the fear of being punished by the demons. When in course of time the mourning runs. its course, the conflict grows less acute, so that the taboo upon the dead is able to diminish in severity or sink into oblivion.

64

I'

• I

-

I

'

TOTEM AND TABOO

Having thus explained the basis of the exceedingly instruc­ tive taboo upon the dead, we must not omit to add a few remarks that may help to increase our understanding of taboo in general. The projection of unconscious hostility on to demons in the case of the taboo upon the dead is only a single instance of a number of processes to which the greatest influence must be attributed in the shaping of the primitive mind. In the case we have been dealing with, projection served the purpose of deal­ ing with an emotional conflict; and it is employed in the same way in a large number of psychical situations that lead to neur­ oses. But projection was not created for the purpose of defence; it also occurs where there is no conflict. The projection out­ wards of internal perceptions is a primitive mechanism, to which, for instance, our sense perceptions are subject, and which therefore normally plays a very large part in determining the form taken by our external world. Under conditions whose nature has not yet been sufficiently established, internal per­ ceptions of emotional and thought processes can be projected outwards in the same way as sense perceptions; they are thus employed for building up the external world, though they should by rights remain part of the internal world. This may have some genetic connection with the fact that the function of attention was originally directed not towards the internal world but towards the stimuli that stream in from the external world, and that that function's only information upon endo­ psychic processes was received from feelings of pleasure and un­ pleasure. It was not until a language of abstract thought had been developed, that is to say, not until the sensory residues of verbal presentations had been linked to the internal processes, that the latter themselves gradually became capable of being perceived. Before that, owing to the projection outwards of in� ternal perceptions, primitive men arrived at a picture of the external world which we, with our intensified conscious percep­ tion, have now to translate back into psychology. [Cf. p. 7¥,] The projection of their own evil impulses into demons is only one portion of a system which constituted the Weltanschauung

II. TABOO AND EMOTIONAL AMBIVALENCE

65

[view of the universe] of primitive peoples, and which we shall come to know as 'animism' in the following essay. There we shall have to -investigate that system's psychological character­ istics, and we shall do so once again by reference to the similar systems which we find constructed by neurotics. For the mo­ ment I will only say that the prototype of all such systems is what we have termed the 'secondary revision' of the content of dreams. 1 And we must not forget that, at and after the stage at which systems are constructed, two sets of reasons can be assigned for every psychical event that is consciously judged­ one set belonging to the system and the other set real but unconscious.1 Wundt ( 1906, 129) remarks that 'among the ·activities attri­ buted by myths all over the world to demons, the harmful pre­ dominate, so that in popular belief bad demons are clearly older than good ones'. It is quite possible that the whole concept of demons was derived from the important relation of the living to the dead. The ambivalence inherent in that relation was ex­ pressed in the subsequent course of human development by the fact that, from the same root, it gave rise to two completely opposed psychical structures: on the one hand fear of demons and ghosts and on the other hand veneration of ancestors.• The fact that demons are always regarded as the spirits of those who have died recent{y shows better than anything the influence of mourning on the origin of the belief in demons. Mourning has a quite specific psychical task to perform: its function is to detach the survivors' memories and hopes from the dead. When this has been achieved, the pain grows less and with it the remorse 1 [Cf. Freud, The Interpretation of Dreams (1900a), Standard Ed., 5, 488 ff.] 3 [Further explained below, p. 94f.] The projected creations of primi­ tive men resemble the personifications constructed by creative writers; for the latter externalize in the form of separate individuals the opposing instinctual impulses struggling within them. 8 In the course of psycho-analyses of neurotics who suffer (or who suffered in their childhood) from fear of ghosts, it is often possible to show without much difficulty that the ghosts are disguises for the patient's parents. Cf. in this connection a paper upon 'Sexual Ghosts' by Haeberlin (1912). Here the person concerned was not the subject's parent (who was dead) but someone else of erotic significance to him.

lI'

66

TOTEM AND TABOO

and self-reproaches and consequently the fear of the demon as well. And the same spirits who to begin with were feared as demons may now expect to meet with friendlier treatment; they are revered as ancestors and appeals are made to them for help.

f•.

'

If we follow the changing relations between survivors·and the dead through the course of ages, it becomes obvious that there has been an extraordinary diminution in ambivalence. It is now quite easy to keep down the unconscious hostility to the dead (though its existence can still be traced) without any particular expenditure of psychical energy. Where, in earlier times, satis­ fied hatred and pained affection fought each other, we now find that a kind of scar has been formed in the shape of piety, which declares 'de mortuis nil nisi bonum'. It is only neurotics whose mourning for the loss of those dear to them is still troubled by obsessive self-reproaches-the secret of which is revealed by psycho-analysis as the old emotional ambivalence. We need not discuss here how this alteration came about or how much share in it is due to a constitutional modification and how much to a real improvement in family relations. But this example suggests the probability that the psychical impulses of primitive peoples were characterized by a higher amount of ambivalence than is to be found in modern civilized man. It is to be supposed that as this ambivalence diminished, taboo ( a symptom of the ambivalence and a compromise between the two co,iflicting impulses) slowly disappeared. Neurotics, who are obliged to reproduce the struggle and the taboo resulting from it, may be said to have inherited an archaic con­ stitution as an atavistic vestige; the need to compensate for this at the behest of civilization is what drives them to their im­ mense expenditure of mental energy. And here we may recall the obscure and puzzling statement by Wundt on the double meaning of the word taboo: 'sacred' and 'unclean'. (See above [pp. 24-5].) Originally, according to him, the word did not possess these two meanings, but described 'what is demonic', 'what may not be touched', thus stressing an important characteristic common to both the extreme concepts. The persistence, however (he added), of this common charac­ teristic was evidence that the ground covered by the two-the

II. TABOO AND EMOTIONAL AMBIVALENCE

67

sacred and the unclean-was originally one and did not become differentiated until later. Our discussions, on the contrary, lead us to the simple con­ clusion that the word 'taboo' had a double meaning from the very first and that it was used to designate a particular kind of ambivalence and whatever arose from it. 'Taboo' is itself an ambivalent word; and one feels on looking back that the well­ attested meaning of the word should alone have made it possible to infer-what has actually been arrived at as a result of exten­ sive researches-that the prohibitions of taboo are to be under­ stood as consequences of an emotional ambivalence. Study of the earliest languages has taught us that there were once many such words, which expressed contrary ideas and in a sense (though not in quite the same sense as the word 'taboo') were ambivalent. 1 Slight modifications in the pronunciation of the antithetical 'primal word' made it possible subsequently to give separate verbal expression to the two contrary ideas which were originally combined in it. The word 'taboo' met with a different fate. As the importance of the ambivalence denoted by it diminished, the word itself, or rather the words analogous to it, fell out of use. I hope to be able, in a later connection, to make it probable that a definite historical chain of events is concealed behind the fate of this concept: that the word was at first attached to certain quite specific human relations which were characterized by great emotional ambivalence, and that its use then spread on to other analogous relations. [See p. 143 ff.l If I am not mistaken, the explanation of taboo also throws light on the nature and origin of conscience. It is possible, with­ out any stretching of the sense of the terms, to speak of a taboo conscience or, after a taboo has been violated, of a taboo sense of guilt. Taboo conscience is probably the earliest form in which the phenomenon of conscience is met with. For what is 'conscience'? On the evidence of language it is related to that of which one is 'most certainly conscious'. 1 Cf. my review of Abel's 'Antithetical Meaning of Primal Words' [Freud, 1910e].

68

TOTEM AND TABOO

Indeed, in some languages the words for 'conscience' and consciousness' can scarcely be distinguished. 1 Conscience is the internal perception of the rejection of a particular wish operating within us. The stress, however, is upon the fact that this rejection has no need to appeal to anything else for support, that it is quite 'certain of itself'. This is even clearer in the case of consciousness of guilt-the perception of the internal condemnation of an act by which we have carried out a particular wish. To put forward any reason for this would seem superfluous: anyone who has a conscience must feel within him the justification for the condemnation, must feel the self­ reproach for the act that has been carried out. This same char­ acteristic is to be seen in the savage's attitude towards taboo. It is a command issued by conscience; any violation of it produces a fearful sense of guilt which follows as a matter of course and of which the origin is unknown. 1 Thus it seems probable that conscience too arose, on a basis of emotional ambivalence, from quite specific human relations to which this ambivalence was attached; and that it arose under the conditions which we have shown to apply in the case of taboo and of obsessional neurosis-namely, that one of the opposing feelings involved shall be unconscious and kept under repression by the compulsive domination of the other one. This conclusion is supported by several things we have learnt from the analysis of neuroses. In the first place, we have found that a feature in the charac­ ter of obsessional neurotics is a scrupulous conscientiousness which is a symptom reacting against the temptation lurking in their unconscious. If their illness becomes more acute, they develop a sense of guilt of the most intense degree. In fact, one

I

I

1 [E.g. the French 'conscience', which has both meanings. The German word for 'conscience' is 'Gewissen', which contains the same root as such words as 'wissen', 'to know', and 'bewusst', 'conscious', as well as the word actually used for comparison in this paragraph and the next, 'gewiss', 'certain' or 'known with certainty'.] 1 The sense of guilt in the case of taboos is not in the least diminished if the violation occurs unwittingly. (Cf. the instances above [p. 42 f.].) An interesting parallel is found in Greek mythology: the guilt of Oedipus was not palliated by the fact that he incurred it without his knowledge and even agamst his intention.

II. TABOO AND EMOTIONAL AMBIVALENCE

69

may venture to say that if we cannot trace the origin of the sense of guilt in obsessional neurotics, there can be no hope of our ever tracing it. This task can be directly achieved in the case of individual neurotic patients, and we may rely upon reaching a similar solution by inference in the case of primitive peoples. In the second place, we cannot help being struck by the fact that a sense of guilt has about it much of the nature of anxiety: we could describe it without any misgivings as 'dread of con­ science'. But the anxiety points to unconscious sources. The psychology of the neuroses has taught us that, if wishful im­ pulses are repressed, their libido is transformed into anxiety. And this reminds us that there is something unknown and un'­ conscious in connection with the sense of guilt, namely the reasons for the act of repudiation. The character of anxiety that is inherent in the sense of guilt corresponds to this unknown factor. 1 Since taboos are mainly expressed in prohibitions, the under­ lying presence of a positive current of desire may occur to us as something quite obvious and calling for no lengthy proofs based on the analogy of the neuroses. For, after all, there is no need to prohibit something that no one desires to do, and a thing that is forbidden with the greatest emphasis must be a thing that is desired. If we were to apply this plausible thesis to our primitive peoples, we should be led to the conclusion that some of their strongest temptations were to kill their kings and priests, to commit incest, to maltreat the dead, and so on-which seems scarcely probable. And we should be met with the most positive contradiction if we were to apply the same thesis to instances in which we ourselves seem most clearly to hear the voice of con­ science. We should maintain with the most absolute certainty that we feel not the slightest temptation to violate any of these prohibitions-the commandment to 'do no murder,, for in­ stance-and that we feel nothing but horror at the notion of violating them. If, however, we were to admit the claims thus asserted by our 1

[It is to be remarked that Freud's views on the origin and nature both of conscience and of anxiety were greatly modified in his later writings. For these later views see Lectures 31 and 32 in his New Intro­ ductory Lectures (1933a).]

70

I

TOTEM AND TABOO

conscience, it would follow, on the one hand, that these prohi­ bitions would be superfluous-both taboo and our own moral prohibitions-and, on the other hand, the fact of conscience would remain unexplained and no place would be left for the relations between conscience, taboo and neurosis. In other words, we should be back in the state of knowledge we were in before we approached the problem from the psycho-analytic ,angle. Suppose, on the other hand, that we were to take into account the finding arrived at by psycho-analysis from the dreams of normal people, to the effect that we ourselves are subject, more strongly and more often than we suspect, to a temptation to kill someone and that that temptation produces psychical effects even though it remains out of sight of our consciousness. Sup­ pose, again, that we were to recognize the compulsive obser­ vances of certain neurotics as being guarantees against an in­ tensified impulse to murder or as being self-punishments on account ofit. In that case we should have to attach still greater importance to our thesis that where there is a prohibition there must be an underlying desire. We should have to suppose that the desire to murder is actually present in the unconscious and that neither taboos nor moral prohibitions are psychologically superfluous but that on the contrary they are explained and justified by the existence ofan ambivalent attitude towards the impulse to murder. One ofthe characteristics ofthis ambivalent relation which I have repeatedly stressed as fundamental-the fact that the posi­ tive current of desire is an unconscious one-opens the way to further considerations and to further possible explanations. Psychical processes in the unconscious are not in every respect identical with those with which our conscious mind is familiar; they enjoy some remarkable liberties that are forbidden to the latter. An unconscious impulse need not have arisen at the point where it makes its appearance; it may arise from some quite other region and have applied originally to quite other persons and connections; it may have reached the place at which it attracts our attention through the mechanism of 'dis­ placement'. Owing, moreover, to the indestructibility and in­ susceptibility to correction which are attributes of unconscious

II. TABOO AND EMOTIONAL AMBIVALENCE

71

processes, it may have survived from very early times to which it was appropriate into later times and circumstances in which its manifestations are bound to seem strange. These are no more than hints, but if they were attentively developed their impor­ tance for our understanding of the growth of civilization would become apparent. Before I conclude this discussion, a further point must not be overlooked which will pave the way for later inquiries. In main­ taining the essential similarity between taboo prohibitions and moral prohibitions, I have not sought to dispute the fact that there must be a psychological difference between them. The only possible reason why the prohibitions no longer take the form of taboos must be some change in the circumstances governing the ambivalence underlying them. In our analytical examination of the problems of taboo we have hitherto allowed ourselves to be led by the points of agree­ ment that we have been able to show between it and obsessional neurosis. But after all taboo is not a neurosis but a social institu­ tion. We are therefore faced with the task of explaining what difference there is in principle between a neurosis and a cultural creation such as taboo. Once again I will take a single fact as my starting-point. It is feared among primitive peoples that the violation of a taboo will be followed by a punishment, as a rule by some serious ill­ ness or by death. The punishment threatens to fall on whoever was responsible for violating the taboo. In obsessional neuroses the case is different. What the patient fears if he performs some forbidden action is that a punishment will fall not on himself but on someone else. This person's identity is as a rule left unstated, but can usually be shown without difficulty by analysis to be one of those closest and most dear to the patient. Here, then, the neurotic seems to be behaving altruistically and the primi­ tive man egoistically. Only if the violation of a taboo is not auto­ matically avenged upon the wrong-doer does a collective feeling arise among savages that they are all threatened by the outrage; and they thereupon hasten to carry out the omitted punish­ ment themselves. There is no difficulty in explaining the mech­ anism of this solidarity. What is in question is fear of an infec-

TOTEM AND TABOO 72 tious example, of the temptation to imitate-that is, of the con­ tagious character of taboo. If one person succeeds in gratifying the repressed desire, the same desire is bound to be kindled in all the other members of the community. In order to keep the temptation down, the envied transgressor must be deprived of the fruit of his enterprise; and the punishment will not infre­ quently give those who carry it out an opportunity of commit­ ting the same outrage under colour of an act of expiation. This is indeed one of the foundations of the human penal s ystem and it is based, no doubt correctly, on the assumption that the pro­ hibited impulses are present alike in the crirninal and in the avenging community. In this, psycho-analysis is no more than confirming the habitual pronouncement of the pious: we are all miserable sinners. How, then, are we to account for the unexpected nobility of mind of the neurotic, who fears nothing on his own account but everything for someone he loves? Analytical inquiry shows that this attitude is not primary. Originally, that is to say at the be­ ginning of the illness, the threat of punishment applied, as in the case of savages, to the patient himself; he was invariably in fear for his own life; it was not until later that the mortal fear was displaced on to another and a loved person. The process is a little complicated, but we can follow it perfectly. At the root of the prohibition there is invariably a hostile impulse against someone the patient loves-a wish that that person should die. This impulse is repressed by a prohibition and the prohibition is attached to some particular act, which, by displacement, represents, it may be, a hostile act against the loved person. There is a threat of death if this act is performed. But the pro­ cess goes further, and the original wish that the loved person may die is replaced by a fear that he may die. So that when the neurosis appears to be so tenderly altruistic, it is merely compen­ sating for an underlying contrary attitude of brutal egoism. We may describe as 'social' the emotions which are determined by showing consideration for another person without taking him as a sexual object. The receding into the back ground of these social factors may be stressed as a fundamental characteristic of the neurosis, though one which is later disguised by over­ compensation.

II. TABOO AND EMOTIONAL AMBIVALENCE

73

I do not propose to linger over the origin of these social im­ pulses and their relation to the other basic human instincts but shall proceed to illustrate the second main characteristic of the neurosis by means of another example. In the forms which it assumes, taboo very closely resembles the neurotic's fear of touching, his 'touching phobia'. Now, in the case ofthe neurosis the prohibition invariably relates to touching of a sexual kind, and psycho-analysis has shown that it is in general true that the instinctual forces that are diverted and displaced in neuroses have a sexual origin. In the case of taboo the prohibited touch­ ing is obviously not to be understood in an exclusively sexual sense but in the more general sense of attacking, of getting con­ trol, and of asserting oneself. If there is a prohibition against touching a chiefor anything that has been in contact with him, this means that an inhibition is to be laid on the same impulse which expresses itselfon other occasions in keeping a suspicious watch upon the chief or even in ill-treating him physically be­ fore his coronation. (See above [p. 49].) Thus the fact which is characteristic of the neurosis is the preponderance of the sexual over the social instinctual elements. The social instincts, however, are them­ selves derived from a combination of egoistic and erotic com­ ponents into wholes of a special kind. This single comparison between taboo and obsessional neur­ osis is enough to enable us to gather the nature of the relation between the different forms ofneurosis and cultural institutions, and to see how it is that the study ofthe psychology ofthe neur­ oses is important for an understanding of the growth ofciviliza­ tion. The neuroses exhibit on the one hand striking and far-reach­ ing points of agreement with those great social institutions, art, religion and philosophy. But on the other hand they seem like distortions of them. It might be maintained that a case of hys­ teria is a caricature ofa work ofart, that an obsessional neurosis is a caricature of a religion and that a paranoic delusion is a caricature of a philosophical system. The divergence resolves itselfultimately into the fact that the neuroses are asocial struc­ tures; they endeavour to achieve by private means what is effected in society by collective effort. Ifwe analyse the instincts at work in the neuroses, we find that the determining influence

74

TOTEM AND TABOO

in them is exercised by instinctual forces of sexual ongm; the corresponding cultural formations, on the other hand, are based upon social instincts, originating from the combination of egoistic and erotic elements. Sexual needs are not capable of uniting men in the same way as are the demands of self-preser­ vation. Sexual satisfaction is essentially the private affair of each individual. The asocial nature of neuroses has its genetic origin in their most fundamental purpose, which is to take flight from an un­ satisfying reality into a more pleasurable world of phantasy. The real world, which is avoided in this way by neurotics, is under the sway of human society and of the institutions collec­ tively created by it. To turn away from reality is at the same time to withdraw from the community of man. 1 1 [Many of the points of comparison in this essay between taboo and obsessional neurosis will be found illustrated in the case history of the 'Rat Man' (1909d). The subject oftaboo was further discussed_ by Freud in 'The Taboo ofVirginity' (1918a).-On the question ofman's'attitude to death, see also 'Thoughts for the Times on War and Death' (1915b). -The remarks on attention on p. 64 may be made clearer by a passage in Chapter VII (D) of The Interpretation ofDreams (1900a), Standard Ed., 5, 574.-The opposition between directly sexual impulsions and social structures is elaborated in Chapter XII (D) ofGroup Psychology (1921c), Standard Ed., 18, 140-2.]

III

ANIMISM, MAGIC AND THE OMNIPOTENCE OF THOUGHTS (1) WRITINGS that seek to apply the findings of psycho-analysis to topics in the field of the mental sciences have the inevitable defect of offering too little to readers of both classes. Such writings can only be in the nature of an instigation: they put before the specialist certain suggestions for him to take into account in his own work. This defect is bound to be extremely evident in an essay which will attempt to deal with the im­ mense domain of what is known as 'animism' • 1 Animism is, in its narrower sense, the doctrine of souls, and, in its wider sense, the doctrine of spiritual beings in general. The term 'animatism' has also been used to denote the theory of the living character of what appear to us to be inanimate ob­ jects [see below, p. 91], and the terms 'animalism' and 'man­ ism' occur as well in this connection. The word 'animism', originally used to describe a particular philosophical system, seems to have been given its present meaning by Tylor.• What led to the introduction of these terms was a realization of the highly remarkable view of nature and the universe adopted by the primitive races of whom we have knowledge, 1 The necessity for a concise treatment of the material involves the omission of any elaborate bibliography. Instead, I will merely refer to the standard works of Herbert Spencer,J. G. Frazer, Andrew Lang, E. B. Tylor and Wilhelm Wundt, from which all that I have to say about animism and magic is derived. My own contribution is visible only in my selection both of material and of opinions. 1 Cf. Tylor (1891,1,425),Wundt (19 6 [142 f. and] 1 3) [and Marett 0 7 (1900,171)]. 75

76

TOTEM AND TABOO

whether in past history or at the present time. They people the world with innumerable spiritual beings both benevolent and malignant; and these spirits and demons they regard as the causes of natural phenomena and they believe that not only animals and plants but all the inanimate objects in the world are animated by them. A third, and perhaps the most important, article of this primitive 'philosophy of nature' 1 strikes us as less strange, since, while we have retained only a very limited belief in the existence of spirits and explain natural phenomena by the agency of impersonal physical forces, we ourselves are not very far removed from this third belief. For primitive peoples believe that human individuals are inhabited by similar spirits. These souls which live in human beings can leave their habitations and migrate into other human beings; they are the vehicle of mental activities and are to a certain extent independent of their bodies. Originally souls were pictured as very similar to persons and only in the course of a long development have they lost their material characteristics and become to a high degree 'spiritualized'. 1 Most authorities incline to the view that these ideas of a soul are the original nucleus of the animistic system, that spirits are only souls that have made themselves independent, and that the souls of animals, plants and objects were constructed on the analogy of human souls. How did primitive men arrive at the peculiar dualistic views on which the animistic system is based? It is supposed that they did so by observing the phenomena of sleep (including dreams) and of death which so much resembles it, and by attempting to explain those states, which are of such close concern to every­ one. The chief starting-point of this theorizing must have been the problem of death. What primitive man regarded as the natural thing was the indefinite prolongation of life-immortal­ ity. The idea of death was only accepted late, and with hesi­ tancy. Even for Q.S it is lacking in content and has no clear con­ notation. There have been very lively but inconclusive discus­ sions upon the part that may have been played in the formation of the basic doctrines of animism by such other observed or 1

1

['Naturphilosophie', the pantheistic philosophy of Schelling.] Wundt (1906), Chapter IV, 'Die Seelenvorstellungen'.

III.ANIMISM, MAGIC, OMNIPOTENCE OF THOUGHTS 77 experienced facts as dream-pictures, shadows, mirror images, and so on.1 It has been regarded as perfectly natural and not in the least puzzling that primitive man should have reacted to the pheno­ mena which aroused his speculations by forming the idea of the soul and then of extending it to objects in the external world.In discussing the fact that the same animistic ideas have emerged among the most various races and at every period, Wundt (1906, 154) declares that 'they are the necessary psychological product of a mythopoeic consciousness ...and in this sense, therefore, primitive animism must be regarded as the spiritual expression of the natural state of man, so far as it.is accessible to our observation'.The justification for attributing life to inanim­ ate objects was already stated by Hume in his Natural History of Religion [Section III]: 'There is an universal tendency among mankind to conceive all beings like themselves, and to transfer to every object those qualities with which they are familiarly acquainted, and of which they are intimately conscious.'• Animism is a system of thought. It does not merely give an explanation of a particular phenomenon, but allows us to grasp the whole universe as a single unity from a single point of view. The human race, if we are to follow the authorities, have in the course of ages developed three such systems of thought-three great pictures of the universe: animistic (or mythological), reli­ gious and scientific.Of these, animism, the first to be created, is perhaps the one which is most consistent and exhaustive and which gives a truly complete explanation of the nature of the universe. This first human Weltanschauung is a psychological theory.It would go beyond our present purpose to show how much of it still persists in modern life, either in the debased form of superstition or as the living basis of our speech, our beliefs and our philosophies. With these three stages in mind, it may be said that animism itself is not yet a religion but contains the foundations on which religions are later built.It is obvious, too, that myths are based 1 Cf. Wundt [loc.cit.], Herbert Spencer [1893, Part I], as well as the general articles in the Encyclopaedia Britannica (1g10-11) on 'Animism', 'Mythology', etc. 1 Quoted by Tylor (1891, 1, 4 7). 7 S,F. XIII-0

78

TOTEM AND TABOO

on animistic premises, though the details of the relation be­ tween myths and animism seem to be unexplained in some essential respects. (2)

d

• I

Our psycho-analytic approach to the subject, however, is from another side. It is not to be supposed that men were in­ spired to create their first system of the universe by pure specu­ lative curiosity. The practical need for controlling the world around them must have played its part. So we are not surprised to learn that, hand in hand with the animistic system, there went a body of instructions upon how to obtain mastery over men, beasts and things-or rather, over their spirits. These in­ structions go by the names of 'sorcery' and 'magic'. 1 Reinach (1905-12, 2, xv) describes them as the 'strategy of animism'; I should prefer, following Hubert and Mauss (1904 [142 ff.]), to regard them as its technique. Can the concepts of sorcery and magic be distinguished? Per­ haps-if we are prepared to show a somewhat arbitrary dis­ regard for the fluctuations of linguistic usage. Sorcery, then, is essentially the art of influencing spirits by treating them in the same way as one would treat men in like circumstances: appeas­ ing them, making amends to them, propitiating them, intimi­ dating them, robbing them of their power, subduing them to one's will-by the same methods that have proved effective with living men. Magic, on the other hand, is something differ­ ent: fundamentally, it disregards spirits and makes use of special procedures and not of everyday psychological methods. It is easy to guess that magic is the earlier and m,.:rP. important branch of animistic technique; for magical methods can, among others, be used in dealing with spirits, 1 and m:.agic can be applied as well in cases where, as it seems to us, the process of spiritualiz­ ing Nature has not yet been carried out. Magic has to serve the most varied purposes-it must subject natural phenomena to the will of man, it must protect the indi['Zauberei' and 'Magie' in the original.] If a spirit is scared away by making a noise and shouting, the action is one purely of sorcery; if compulsion is applied to it by getting hold of its name, magic has been used against it. 1

2

HI. ANIMISM, MAGIC, OMNIPOTENCE OF THOUGHTS

79

vidual from his enemies and from dangers and it must give him power to injure his enemies. But the principle on the presump­ tion of which magical action is bas�d-or, more properly, the principle of magic-is so striking that none of the authorities has failed to recognize it. Tylor [1891, 1, u6], if we leave on one side an accompanying moral judgement, states it in its most suc­ cinct form as mistaking an ideal connection for a real one. I will illustrate this feature from two groups of magical acts. One of the most widespread magical procedures for injuring an enemy is by making an effigy of him from any convenient material. Whether the effigy resembles him is of little account: any object can be 'made into' an effi gy of him. Whatever is then done to the effigy, the same thing happens to the detested orig­ inal; whatever part of the farmer's body is damaged, the same part of the latter's becomes diseased. The same magical tech­ nique may be employed, not only for purposes of private en­ mity, but also for pious ends and for giving help to gods against malignant demons. I will quote from Frazer (19ua, 1, 67): 'Every night when the sun-god Ra sank down to his home in the glowing west he was assailed by hosts of demons under the leadership of the arch-fiend Apepi. All night long he fought them, and sometimes by day the powers of darkness sent up clouds even into the blue Egyptian sky to obscure his light and weaken his power. To aid the sun-god in this daily struggle, a ceremony was daily performed in his temple at Thebes. A figure of his foe Apepi, represented as a crocodile with a hideous face or a serpent with many coils, was made of wax, and on it the demon's name was written in green ink. Wrapt in a papyrus case, on which another likeness of Apepi had been drawn in green ink, the figure was then tied up with black hair, spat upon, hacked with a stone knife, and cast on the ground. There the priest trod on it with his left foot again and again, and then burnt it in a fire made of a certain plant or grass. When Apepi himself had thus been effectually disposed of, waxen effigies of each of his principal demons, and of their fathers, mothers and children, were made and burnt in the same way. The service, accompanied by the recitation of certain prescribed spells, was repeated not merely morning, noon and night, but whenever a

80

''

• I

I

I

TOTEM AND TABOO

storm was raging, or heavy rain had set in, or black clouds were stealing across the sky to hide the sun's bright disc. The fiends of darkness, clouds, and rain felt the injuries inflicted on their images as if they had been done to themselves; they passed away, at least for a time, and the beneficent sun-god shone out triumphant once more.' 1 From the vast number of magical acts having a similar basis I will only draw attention to two more, which have played a large part among primitive peoples of every age and which per­ sist to some degree in the myths and cults of higher stages of civilization-that is, rituals for producing rain and fertility. Rain is produced magically by imitating it or the clouds and storms which give rise to it, by 'playing at rain', one might almost say. In Japan, for instance, 'a party of Ainos will scatter water by means of sieves, while others will take a porringer, fit it up with sails and oars as if it were a boat, and then push or draw it about the village and gardens'.• In the same way, the fertility of the earth is magically promoted by a dramatic representation of human intercourse. Thus, to take one from a countless number of instances, 'in some parts of Java, at the season when the bloom will soon be on the rice, the husbandman and his wife visit their fields by night and there engage in sexual inter­ course' to encourage the fertility of the rice by their example.• There is a dread, however, that prohibited, incestuous sexual relations may cause a failure of the crops and make the earth sterile.' Certain negative observances, that is, magical precautions, must be included in this first group. 'When a Dyak village has turned out to hunt wild pigs in the jungle, the people who stay at home may not touch oil or water with their hands during the absence of their friends; for if they did so, the hunters would all be "butter-fingered" and the prey would slip through their 1 It seems probable that the biblical prohibition against making an image of any living thing originated, not from any objection to the plas­ tic arts, but from a desire to deprive magic (which was abominated by the Hebrew religion) of one of its tools. Cf. Frazer (1911a, 1, 87 n.). 1 [Frazer (1911a, 1, 251), quoting Batchelor (1901, 333).] 8 Frazer (1911a, 2, 98) [quoting Wilken (1884, 958)]. 'An echo of this is to be found in the Oedipus Rex of Sophocles [e.g. in the prologue and first Chorus].

,...

III. ANIMISM, MAGIC, OMNIPOTENCE OF THOUGHTS

81

1

hands.' Or again, 'while a Gil yak hunter is pursuing game in the forest, his children at home are forbidden to make drawings on wood or on sand; for they fear that if the children did so, the paths in the forest would become as perplexed as the lines in the drawings, so that the hunter might lose his way and never return.'• In these last, as in so many other instances of the workings of magic, the element of distance is disregarded; in other words, telepathy is taken for granted. We shall find no difficulty, there­ fore, in understanding this characteristic of magic. There can be no doubt what is to be regarded as the operative factor in all these examples. It is the similar#y between the act performed and the result expected. For this reason Frazer de­ scribes this sort of magic as 'imitative' or 'homoeopathic'. If I wish it to rain, I have only to do something that looks like rain or is reminiscent of rain. At a later stage of civilization, instead of this rain-magic, processions will be made to a temple and prayers for rain will be addressed to the deity living in it. Finally, this religious technique will in its turn be given up and attempts will be made to produce effects in the atmosphere which will lead to rain. In a second group of magical acts the principle of similarity plays no part, and its place is taken by another one, the nature of which will at once become clear from the following examples. There is another procedure by which an enemy can be in­ jured. One gets possession of some of his hair or nails or other waste products or even a piece of his clothing, and treats them in some hostile way. It is then exactly as though one had got possession of the man himself; and he himself experiences what­ ever it is that has been done to the objects that originated from him. In the view of primitive man, one of the most important parts of a person is his name. So that if one knows the name of a man or of a spirit, one has obtained a certain amount of power over the owner of the name. This is the origin of the remarkable precautions and restrictions in the use of names which we have already touched upon in the essay on taboo. (See p. 54 ff.) In 1 Frazer Frazer

8

(19ua, (1911a,

1, H!O) [quoting Roth (1896, 1, 430)]. 1, 122) [quoting Labbe (1903, 268)].

82

TOTEM AND TABOO

these examples the place of similarity is evidently taken by affinity. The higher motives for cannibalism among primitive races have a similar origin. By incorporating parts of a person's body through the act of eating, one at the same time acquires the qualities possessed by him. This leads in certain circumstances to precautions and restrictions in regard to diet. A woman who is with child will avoid eating the flesh of certain animals for fear that any undesirable qualities they may have (cowardice, for instance) might be passed over to the child that is nourished by her. The magical power is not affected even if the connection between the two objects has already been severed or even if the contact occurred only on a single important occasion. For in­ stance, the belief that there is a magical bond between a wound and the weapon which caused it may be traced unaltered for thousands of years. If a Melanesian can obtain possession of the bow which caused his wound, he will keep it carefully in a cool place so as to reduce the inflammation of the wound. But if the bow was left in the enemy's possession, it will undoubtedly be hung up close to the fire so that the wound may become thor­ oughly hot and inflamed. 1 Pliny (in his Natural History, Book xxviii [Chapter 7]) tells us that 'if you have wounded a man and are sorry for it, you have only to spit on the hand that gave the wound, and the pain of the sufferer will be instantly alleviated'. [Frazer, 19ua, 1, 201.] Francis Bacon (in his Sylva Sylvarum [X, § 998]) mentions that 'it is constantly received and avouched that the anointing of the weapon that maketh the wound will heal the wound itself'. English country people are said even to­ day to follow this prescription, and if they cut themselves with a scythe carefully keep the instrument clean, to prevent the wound from festering. 'At Norwich in June 1902 a woman named Matilda Henry accidentally ran a nail into her foot. Without examining the wound, or even removing her stocking, she caused her daughter to grease the nail, saying that if this were done no harm would come of the hurt. A few days after­ wards she died of lockjaw'-as a result of this displaced anti­ sepsis. (Frazer, ibid., 203.) The last group of instances exemplify what Frazer distin1 [Frazer (1911a, 1, 201), quoting Codrington (1891, 310).]

III. ANIMISM, MAGIC, OMNIPOTENCE OF THOUGHTS

83

guishes from 'imitative' magic under the name of 'contagious' magic. What is believed to be their effective principle is no longer similarity but spatial connection, contiguity, or at least imagined contiguity-the recollection ofit. Since,however,simi­ larity and contiguity are the two essential principles of pro­ cesses of association, it appears that the true explanation of all the folly ofmagical observances is the domination ofthe associ­ ation of ideas. The aptness of Tylor's description of magic which I have already quoted [p. 79] now becomes evident: mistaking an ideal connection for a real one. Frazer (1911a, 1,420) has put it almost in the same words: 'Men mistook the order of their ideas for the order ofnature, and hence imagined that the control which they have, or seem to have, over their thoughts, permitted them to exercise a corresponding control over things.' We shall at first be surprised to learn that this illuminating explanation of magic has been rejected by some writers as un­ satisfactory (e.g. Thomas, 1910-11a). On reflection, however, it will be seen that the criticism is justified. The associative theory ofmagic merely explains the paths along which magic proceeds; it does not explain its true essence, namely the misunderstand­ ing which leads it to replace the laws ofnature by psychological ones. Some dynamic factor is evidently missing. But whereas the critics of Frazer's theory have gone astray in their search for it, it will be easy to arrive at a satisfactory explanation of magic merely by carrying the associative theory further and deeper. Let us consider first the simpler and more important case of imitative magic. According to Frazer (19ua, 1, 54) it can be practised by itself, whereas contagious magic as a rule presup­ poses the other. It is easy to perceive the motives which lead men to practise magic: they are human wishes. All we need to suppose is that primitive man had an immense belief in the power of his wishes. The basic reason why what he sets about by magical means comes to pass is, after all, simply that he wills it. To begin with, therefore, the emphasis is only upon his wish. Children are in an analogous psychical situation, though their motor efficiency is still undeveloped. I have elsewhere

TOTEM AND TABOO

84

(19ub) put forward the hypothesis that, to begin with, they satisfy their wishes in a hallucinatory manner, that is, they create a satisfying situation by means of centrifugal excitations oftheir sense organs. An adult primitive man has an alternative method open to him. His wishes are accompanied by a motor impulse, the will, which is later destined to alter the whole face oftp.e earth i,n order to satisfy his wishes. This motor impulse is at first employed to give a representation of the satisfying situ­ ation in such a way that it becomes possible to experience the satisfaction by means ofwhat might be described as motor hal­ lucinations. This kind of representation of a satisfied wish is quite comparable to children's play, which succeeds their earlier purely sensory technique of satisfaction. If children and primi­ tive men find play and imitative representation enough for them, that is not a sign of their being unassuming in our sense or oftheir resignedly accepting their actual impotence. It is the easily understandable result of the paramount virtue they ascribe to their wishes, of the will that is associated with those wishes and of the methods by which those wishes operate. As time goes on, the psychological accent shifts from the motives for the magical act on to the measures by which it is carried out­ that is, on to the act itself. (It would perhaps be more correct to say that it is only these measures that reveal to the subject the excessive valuation which he attaches to his psychical acts.) It thus comes to appear as though it is the magical act itself which, owing to its similarity with the desired result, alone determines the occurrence of that result. There is no opportun­ ity, at the stage ofanimistic thinking, for showing any objective evidence of the true state ofaffairs. But a possibility ofdoing so does arrive ata later time, when, though all of these procedures are still being carried out, the psychical phenomenon of doubt has begun to emerge as an expression of a tendency to repres­ sion. At that point, men will be ready to admit that conjuring up spirits has no result unless it is accompanied by faith, and that the magical power of prayer fails if there is no piety at work behind it. 1 1

Cf. the King in Hamlet (m. 3): My words fly up, my thoughts remain below: Words without thoughts never to heaven go.

III. ANIMISM, MAGIC, OMNIPOTENCE OF THOUGHTS

85

The fact that it has been possible to construct a system of con­ tagious magic on associations of contiguity shows that the im­ portance attached to wishes and to the will has been extended from them on to all those psychical acts which are subject to the will. A general overvaluation has thus come about of all men­ tal processes-an attitude towards the world, that is, which, in view of our knowledge of the relation between reality and thought, cannot fail to strike us as an overvaluation of the latter. Things become less important than ideas of things: what­ ever is done to the latter will inevitably also occur to the former. Relations which hold between the ideas of things are assumed to hold equally between the things themselves. Since distance is of no importance in thinking-since what lies furthest apart both in time and space can without difficulty be comprehended in a single act of consciousness-so, too, the world of magic has a telepathic disregard for spacial distance and treats past situ­ ations as though they were present. In the animistic epoch the reflection of the internal world is bound to blot out the other picture of the world-the one which we seem to perceive. It is further to be noticed that the two principles of association --similarity and contiguity-are both included in the more comprehensive concept of 'contact'. Association by contiguity is contact in the literal sense; association by similarity is contact in the metaphorical sense. The use of the same word for the two kinds of relation is no doubt accounted for by some identity in the psychical processes concerned which we have not. yet grasped. We have here the same range of meaning of the idea of 'contact' as we found in our analysis of taboo. (C£ y. 27.) By way of summary, then, it may be said that the principle governing magic, the technique of the animistic mode of think­ ing, is the principle of the 'omnipotence of thoughts'.

(3) I have adopted the term 'omnipotence of thoughts' from a highly intelligent man who suffered from obsessional ideas an_d who, after having been set right by psycho-analytic treatment, was able to give evidence of his efficiency and good sense.

86

I' '

TOTEM AND TABOO

(Cf. Freud, 1909d. 1) He had coined the phrase as an explanation of all the strange and uncanny events by which he, like others afflicted with the same illness, seemed to be pursued. If he thought of someone, he would be sure to meet that very person immediately afterwards, as though by magic. If he suddenly asked after the health of an acquaintance whom he had not seen for a long time, he would hear that he had just died, so that it would look as though a telepathic message had arrived from him. If, without any really serious intention, he swore at some stranger, he might be sure that the man would die soon after­ wards, so that he would feel responsible for his death. In the course of the treatment he himself was able to tell me how the deceptive appearance arose in most of these cases, and by what contrivances he himself had helped to strengthen his own super­ stitious beliefs. All obsessional neurotics are superstitious in this way, usually against their better judgement.• It is in obsessional neuroses that the survival of the omnipo­ tence of thoughts is most clearly visible and that the conse­ quences of this prixnitive mode of thinking come closest to con­ sciousness. But we must not be xnisled into supposi�g that it is a distinguishing feature of this particular neurosis, for analytic investigation reveals the same thing in the other neuroses as well. In all of them what deterxnines the formation of symptoms is the reality not of experience but of thought. Neurotics live in a world apart, where, as I have said elsewhere [1911b, near the end of the paper], only 'neurotic currency' is legal tender; that is to say, they are only affected by what is thought with intensity and pictured with emotion, whereas agreement with external reality is a matter of no importance. What hysterics repeat in their attacks and fix by means of their symptoms are experiences which have occurred in that fori:n only in their imagination­ though it is true that in the last resort those imagined experi­ ences go back to actual events or are based upon them. To [This was the 'Rat Man'. Cf. Standard Ed., 10, 233 ff.] We appear to attribute an 'uncanny' quality to impressions that seek to confirm the omnipotence of thoughts and the animistic mode of thinking in general, after we have reached a stage at which, in our judgement, we have abandoned such beliefs. [Cf. Freud's subsequent paper on 'The Uncanny' (1919h).] 1

1

I

I' III. ANIMISM, MAGIC, OMNIPOTENCE OF THOUGHTS

87

attribute the neurotic sense of guilt to real misdeeds would show an equal misunderstanding. An obsessional neurotic may be weighed down by a sense of guilt that would be appropriate in a mass-murderer, while in fact, from his childhood onwards, he has behaved to his fellow-men as the most considerate and scrupu­ lous member of society. Nevertheless, his sense of guilt has a justification: it is founded on the intense and frequent death­ wishes against his fellows which are unconsciously at work in him. It has a justification if what we take into account are un­ conscious thoughts and not intentional deeds. Thus the omni­ potence of thoughts, the overvaluation of mental processes as compared with reality, is seen to have unrestricted play in the emotional life of neurotic patients and in everything that derives from it. If one of them undergoes psycho-analytic treatment, which makes what is unconscious in him conscious, he"Will be unable to believe that thoughts are free and will constantly be afraid of expressing evil wishes, as though their expression would lead inevitably to their fulfilment. This behaviour, as well as the superstitions which he practises in ordinary life, reveals his re­ semblance to the savages who believe they can alter the external world by mere thinking. The primary obsessive acts of these neurotics are of an en­ tirely magical character. If they are not charms, they are at all events counter-charms, designed to ward off the expectations of disaster with which the neurosis usually starts. Whenever I have succeeded in penetrating the mystery, I have found that the ex­ pected disaster was death. Schopenhauer has said that the prob­ lem of death stands at the outset of every philosophy; and we have already seen [p. 76] that the origin of the belief in souls and in demons, which is the essence of animism, goes back to the impression which is made upon men by death. It is difficult to judge whether the obsessive or protective acts performed by obsessional neurotics follow the law of similarity (or, as the case may be, of contrast); for as a rule, owing to the prevailing con­ ditions of the neurosis, they have been distorted by being dis­ placed on to something very small, some action in itself of the greatest triviality. 1 The protective formulas of obsessional 1 A further motive for such displacement on to a very small action will appear in what follows. (Cf. Standard Ed., 10, 241.]

:I

88

TOTEM AND TABOO

neuroses, too, have their counterpart in the formulas of magic. It is possible, however, to describe the course of development of obsessive acts: we can show how they begin by being as remote as possible from anything sexual-magical defences against evil wishes-and how they end by being substitutes for the forbidden sexual act and the closest possible imitations of it.

>

I

I' I! 1,1' '

'I

.Ii

If we are prepared to accept the account given above of the evolution of human views· of the universe-an animistic phase followed by a religious phase and this in turn by a scientific one -it will not be difficult to follow the vicissitudes of the 'omni­ potence of thoughts' through these different phases. At the animistic stage men ascribe omnipotence to themselves. At the religious stage they transfer it to the gods but do not seriously abandon it themselves, for they reserve the power of influencing the gods in a variety of ways according to their wishes. The scientific view of the universe no longer affords any room for human omnipotence; men have acknowledged their smallness and submitted resignedly to death and to the other necessities of nature. None the less some of the primitive belief in omnipo­ tence still survives in men's faith in the power of the human mind, which grapples with the laws of reality. If we trace back the development of libidinal trends as we find them in the individual from their adult forms to the first begin­ nings in childhood, an important distinction emerges, which I have described in my Three Essays on the Theory of Sexuality ( I 905d). Manifestations of the sexual instincts can be observed from the very first, but to begin with they are not yet directed towards any external object. The separate instinctual compon­ ents of sexuality work independently of one another to obtain pleasure and find satisfaction in the subject's own body. This stage is known as that of auto-erotism and it is succeeded by one in which an object is chosen. Further study has shown that it is expedient and indeed in­ dispensable to insert a third stage between these two, or, putting it in another way, to divide the first stage, that of auto-erotism, into two. At this intermediate stage, the importance of which is being made more and more evident by research, the hitherto isolated sexual instincts have already come together into a single

... III. ANIMISM, MAGIC, OMNIPOTENCE OF THOUGHTS

89

whole and have also found an object. But this object is not an external one, extraneous to the subject, but it is his own ego, which has been constituted at about this same time. Bearing in mind pathological fixations of this new stage, which become observable later, we have given it the name of 'narcissism'. The subject behaves as though he were in love with himself; his ego­ istic instincts and his libidinal wishes are not yet separable under our analysis. Although we are not yet in a position to describe with suffi­ cient accuracy the characteristics of this narcissistic stage, at which the hitherto dissociated sexual instincts come together into a single unity and cathect the ego as an object, we suspect already that this narcissistic organization is never wholly aban­ doned. A human being remains to some extent narcissistic even after he has found external objects for his libido. The cathexes of objects which he effects are as it were emanations of the libido that still remains in his ego and can be drawn back into it once more. The state of being in love, which is psychologically so remarkable and is the normal prototype of the psychoses, shows these emanations at their maximum compared to the level of self-love. 1 Primitive men and neurotics, as we have seen, attach a high valuation-in our eyes an over-valuation-to psychical acts. This attitude may plausibly be brought into relation with narcissism and regarded as an essential component of it. It may be said that in primitive men the process of thinking is still to a great extent sexualized. This is the origin of their belief in the omni­ potence of thoughts, their unshakable confidence in the possi­ bility of controlling the world and their inaccessibility to the experiences, so easily obtainable, which could teach them man's true position in the universe. As regards neurotics, we find that on the one hand a considerable part of this primitive attitude has survived in their constitution, and on the other hand that the sexual repression that has occurred in them has brought about a further sexualization of their thinking processes. The psychological results must be the same in both cases, whether the libidinal hypercathexis of thinking is an original one or has 1 [The wh ole question of narcissism was discussed at length by Freu d later on in his paper 'On Narcissism: an Intr oduction' (1914,C).]

\1 /,

I

90

.I

TOTEM AND TABOO

been produced by regression: intellectual narcissism and the omnipotence of thoughts. 1 If we may regard the existence among primitive races of the omnipotence of thoughts as evidence in favour of narcissism, we are encouraged to attempt a comparison between the phases in the development of men's view of the universe and the stages of an individual's libidinal development. The animistic phase would correspond to narcissism both chronologically and in its content; the religious phase would correspond to the stage of object-choice of which the characteristic is a child's attachment to his parents; while the scientific phase would have an exact counterpart in the stage at which an individual has reached maturity, has renounced the pleasure principle, adjusted him­ self to reality and turned to the external world for the object of his desires. 1 In only a single field of our civilization has the omnipotence of thoughts been retained, and that is in the field of art. Only in art does it still happen that a man who is consumed by desires performs something resembling the accomplishment of those desires and that what he does in play produces emotional effects -thanks to artistic illusion-just as though it were something real. People speak with justice ofthe 'magic ofart' and compare artists to magicians. But the comparison is perhaps more signi­ ficant than it claims to be. There can be no doubt that art did not begin as art for art's sake. It worked originally in the service of impulses which are for the most part extinct to-day. And among them we may suspect the presence of many magical purposes.8 1

'It is almost an axiom with writers on this subject, that a sort of Solipsism, or Berkleianism (as Professor Sully terms it as he finds it in the Child), operates in the savage to make him refuse to recognize death as a fact.' (Marett, 1900, 178.) 1 I will only briefly allude here to the fact that the original narcissism of children has a decisive influence upon our view of the development of their character and excludes the possibility of their having any primary sense of inferiority [as Alder suggested (see Part III of Freud, 1914d)]. 8 Cf. Reinach, 'L'art et la magie' (1905-12, 1, 125-36). In Reinach's opinion the primitive artists who left behind the carvings and paintings of animals in the French caves, did not desire to 'please' but to 'evoke' or conjure up. He thus explains why it is that these pictures are situated in the darkest and most inaccessible parts of the caves and that dangerous

III. ANIMISM, MAGIC, OMNIPOTENCE OF THOUGHTS

91

(4)

Thus the first picture which man formed of the world-anim­ ism-was a psychological one. It needed no scientific basis as yet, since science only begins after it has been realized that the world is unknown and that means must therefore be sought for getting to know it. Animism came to primitive man naturally and as a matter of course. He knew what things were like in the world, namely just as he felt himself to be. We are thus prepared to find that primitive man transposed the structural conditions of his own mind 1 into the external world; and we may attempt to reverse the process and put back into the human mind what animism teaches as to the nature of things. The technique of animism, magic, reveals in the clearest and most unmistakable way an intention to impose the laws govern­ ing mental life upon real things; in this, spirits need not as yet play any part, though spirits may be taken as objects of magical treatment. Thus the assumptions of magic are more fundamental and older than the doctrine of spirits, which forms the kernel of animism. Our psycho-analytic point of view coincides here with a theory put forward by R.R. Marett (1900), who postulates a pre-animistic stage before animism, the character of which is best indicated by the term 'animatism', the doctrine of the uni­ versality of life. Experience has little light to throw on pre­ animism, since no race has yet been discovered which is without the concept of spirits. (Cf. Wundt, 1906, 171 ff.) Whereas magic still reserves omnipotence solely for thoughts, beasts of prey do not appear among them. 'Les modernes parlent souvent, par hyperbole, de la magie du pinceau ou du ciseau d'un grand artiste et, en general, de la magie de l'art. Entendu au sens propre, qui est celui d'une contrainte mystique exercee par la volonte de l'homme sur d'autres volontes ou sur les choses, cette expression n'est plus ad­ missible; mais nous avons vu qu'elle etait autrefois rigoureusement vraie, du moins dans !'opinion des artistes.' (Ibid., 136.) ['In modern times people often speak metaphorically of the magic of a great artist's brush or chisel, or more generally of the magic of art. This expression is no longer permissible in its proper sense of a mystical force brought to bear by the human will upon other wills or upon objects; but, as we have seen, there was a time when it was literally true-at least in the artists' opinion.'] 1 Which he was aware of by what is known as endopsychic perception.

92

TOTEM AND TABOO

animism hands some of it over to spirits and so prepares the way for the construction of a religion. What, we may ask, can have induced a primitive man to make this first act of renunciation? It can scarcely have been a recognition of the falseness of his premises, for he continued to practise the magical technique.

1 l•

Spirits and demons, as I have shown in the last essay, are only projections of man's own emotional impulses. 1 He turns his emotional cathexes into persons, he peoples the world with them and meets his internal mental processes again outside him­ self-in just the same way as that intelligent paranoic, Schre­ ber, found a reflection of the attachments and detachments of his libido in the vicissitudes of his confabulated 'rays of God'.• I propose to avoid (as I have already done elsewhere8) enter­ ing into the general problem of the origin of the tendency to pro­ ject mental processes into the outside. It is, however, safe to assume that that tendency will be intensified when projection promises to bring with it the advantage of mental relief. Such an advantage may be expected with certainty where a conflict has arisen between different impulses all of which are striving to­ wards omnipotence-for they clearly cannot all become omni­ potent. The pathological process in paranoia in fact makes use of the mechanism of projection in order to deal with mental con­ flicts of this kind. The typical case of such a conflict is one be­ tween the two members of a pair of opposites-the case of an ambivalent attitude, which we have examined in detail as it appears in someone mourning the death of a loved relative. [Cf. p. 60 ff.] This kind of case must seem particularly likely to provide a motive for the creation of projections. Here again we are in agreement with the writers who maintain that the first­ born spirits were evil spirits, and who derive the idea of a soul from the impression made by death upon the survivors. The only difference is that we do not lay stress on the intellectual prob­ lem with which death confronts the living; in our view the force 1

I assume that at this early narcissistic stage cathexes arising from libidinal and from other sources of excitation may still be indistinguish­ able from one another. 1 Cf. Schreber (1903) and Freud (1911c [Section III (4)]). 1 In my paper on Schreber (Freud, 1911c [Section III]).

Ill. ANIMISM, MAGIC, OMNIPOTENCE OF THOUGHTS

93

which gives the impetus to research is rather to be attributed to the emotional conflict into which the survivors are plunged. Thus man's first theoretical achievement-the creation of spirits-seems to have arisen from the same source as the first moral restrictions to which he was subjected-the observances of taboo. The fact that they had the same origin need not imply, however, that they arose simultaneously. If the survivors' posi­ tion in relation to the dead was really what first caused primi­ tive man to reflect, and compelled him to hand over some of his omnipotence to the spirits and to sacrifice some of his freedom of action, then these cultural products would constitute a first acknowledgment of 'Av